Repositorio Institucional de Documentos

2021 126 Lorenzo Vallan Fluorescent polymeric carbon dots: from synthesis and elucidation of chemical struture towards photoactive hybrid materials Director/es Benito Moraleja, Ana M. Maser, Wolfgang K. © Universidad de Zaragoza Servicio de Publicaciones ISSN 2254-7606 Lorenzo Vallan FLUORESCENT POLYMERIC CARBON DOTS: FROM SYNTHESIS AND ELUCIDATION OF CHEMICAL STRUTURE TOWARDS PHOTOACTIVE HYBRID MATERIALS Director/es Benito Moraleja, Ana M. Maser, Wolfgang K. Tesis Doctoral Autor 2019 Repositorio de la Universidad de Zaragoza – Zaguan http://zaguan.unizar.es UNIVERSIDAD DE ZARAGOZA Escuela de Doctorado Programa de Doctorado en Ingeniería Química y del Medio Ambiente FLUORESCENT POLYMERIC CARBON DOTS: FROM SYNTHESIS AND ELUCIDATION OF CHEMICAL STRUCTURE TOWARDS PHOTOACTIVE HYBRID MATERIALS TESIS DOCTORAL Lorenzo Vallan 2018 CONSEJO SUPERIOR DE INVESTIGACIONES CIENTÍFICAS (CSIC) INSTITUTO DE CARBOQUÍMICA FLUORESCENT POLYMERIC CARBON DOTS: FROM SYNTHESIS AND ELUCIDATION OF CHEMICAL STRUCTURE TOWARDS PHOTOACTIVE HYBRID MATERIALS Memoria presentada en el marco del Programa de Doctorado de Ingeniería Química y Tecnologías del Medio Ambiente de la Universidad de Zaragoza, para optar al grado de Doctor por: Lorenzo Vallan Diciembre 2018 Directores: Ana Maria Benito Moraleja Wolfgang Maser Ana María Benito Moraleja y Wolfgang Maser Investigadores Científicos del Consejo Superior de Investigaciones Científicas CERTIFICAN Que la Memoria, titulada “Preparation and investigation of fluorescent carbon dots and carbon dots-based materials” ha sido realizada bajo nuestra dirección en el Instituto de Carboquimica de Zaragoza (CSIC) por D. Lorenzo Vallan, autorizando su presentación. Y para que así conste, firmamos el presente certificado en Zaragoza, a 12 de diciembre de 2018 Dr. Wolfgang Maser Dra. Ana María Benito Moraleja iv Content Aknowledgement ........................................................................................................................ i List of Figures .......................................................................................................................... vii List of Schemes ........................................................................................................................ xx List of Tables ........................................................................................................................... xxi List of abbreviations and acronyms ...................................................................................... xxiii 1. LITERATURE OVERVIEW .......................................................................................... 1 1.1. Definition of carbon dots ......................................................................................... 2 1.2. Understanding CDs chemical nature and optical properties: a chronological approach .............................................................................................................................. 3 1.3. Precursors and synthesis methods of CDs ............................................................. 14 1.4. Quenching of the CDs fluorescence....................................................................... 18 1.5. Toxicity of CDs ...................................................................................................... 22 1.6. Functionalization of CDs for biomedical applications .......................................... 23 1.7. CDs in photocatalysis and solar cells ..................................................................... 28 1.8. References .............................................................................................................. 30 2. ELUCIDATION OF THE RELATIONSHIP BETWEEN POLYMER STRUCTURE AND BLUE FLUORESCENCE OF CARBON DOTS ....................................................... 38 2.1. Abstract .................................................................................................................. 39 2.2. Introduction ............................................................................................................ 39 2.3. Experimental Section ............................................................................................. 40 2.4. Results and discussion ........................................................................................... 45 2.5. Conclusion ............................................................................................................. 61 2.6. References .............................................................................................................. 62 3. A VERSATILE METHOD FOR THE CONTROLLABLE ROOM-TEMPERATURE SYNTHESIS AND IN-SITU FUNCTIONALIZATION OF FLUORESCENT CARBON DOTS .................................................................................................................................... 65 3.1. Abstract .................................................................................................................. 66 3.2. Introduction ............................................................................................................ 66 3.3. Experimental section .............................................................................................. 67 3.4. Results and discussion ........................................................................................... 68 3.5. Conclusion ............................................................................................................. 75 3.6. References .............................................................................................................. 75 v 4. ELECTRONIC INTERACTIONS IN CDs/MoS2 ELECTROSTATIC COMPLEX .... 77 4.1. Abstract .................................................................................................................. 78 4.2. Introduction ............................................................................................................ 78 4.3. Experimental section .............................................................................................. 79 4.4. Titration experiment ............................................................................................... 89 4.5. Conclusions ............................................................................................................ 94 4.6. References .............................................................................................................. 95 5. ELECTRONIC INTERACTIONS IN COVALENT CDs-TMDs HYBRIDS .............. 96 5.1. Abstract .................................................................................................................. 97 5.2. Introduction ............................................................................................................ 97 5.3. Experimental section .............................................................................................. 98 5.4. Results and discussion ......................................................................................... 104 5.5. Conclusions .......................................................................................................... 119 5.6. References ............................................................................................................ 120 6. GENERAL CONCLUSION AND OUTLOOK ......................................................... 122 6.1. General conclusions ............................................................................................. 123 6.2. Outlook ................................................................................................................ 125 6.3. Conclusiones generales ........................................................................................ 127 6.4. Perspectiva ........................................................................................................... 127 7. ANNEX A ................................................................................................................... 129 7.1. Calculation of the hydrodynamic radius .............................................................. 132 7.2. Calculation of the Quantum Yield ....................................................................... 133 8. ANNEX B ................................................................................................................... 134 8.1. Abstract ................................................................................................................ 135 8.2. Synthesis of CDsA, CDsB and CDsC .................................................................. 135 8.3. Size determination of CDsA-C ............................................................................ 136 8.4. Optical properties of CDsA-C.............................................................................. 139 8.5. Structure characterization of CDsA-C ................................................................. 141 8.6. Conclusion ........................................................................................................... 143 8.7. Details of the DFT and TDDFT calculations ....................................................... 143 8.8. References ............................................................................................................ 164 9. ANNEX C ................................................................................................................... 165 9.1. Abstract ................................................................................................................ 166 vi 9.2. Experimental section ............................................................................................ 166 9.3. Characterization ................................................................................................... 173 10. ANNEX D ............................................................................................................... 236 10.1. Abstract ............................................................................................................ 237 10.2. Experimental section ........................................................................................ 237 10.3. Results and discussion ...................................................................................... 237 11. ANNEX E ................................................................................................................ 239 11.1. Abstract ............................................................................................................ 240 11.2. Experimental part ............................................................................................. 240 11.3. Results and discussion ...................................................................................... 242 11.4. Conclusions ...................................................................................................... 246 LIST OF SCIENTIFIC CONTRIBUTIONS .......................................................................... 247 vii List of Figures Figure 1.1. number of publications containing the words “carbon dots” in the title from 2008 to 2018, source: Scopus). ........................................................................................................... 4 Figure 1.2. (left) TEM images of nanocrystals and other carbonaceous material, (right) absorption and emission of the nanocrystals, from ref. 6). ......................................................... 4 Figure 1.3. PL emission of PEG1500N passivated CDs, excited at different wavelengths (from ref. 9). .......................................................................................................................................... 5 Figure 1.4. (a) PL due to the quantum size effect. The sum of different conjugated domains with distinct energy band gaps can be responsible for the observed excitation-dependent PL behavior. (b) Oxygen groups located at the edges of the conjugated domains can act as localized energy traps, which promote the radiative relaxation. ................................................ 7 Figure 1.5. reactions that occur during the hydrothermal treatment of citric acid and ethylenediamine towards the formation of molecular fluorophores, polymer clusters and the carbonized core (from ref. 30) (b) IPCA fluorophore and its optical properties (from ref. 30). (c) molecular fluorophore and core contributions to the absorption of CDs (from ref 36). (d) the hydrothermal treatment of citric acid with different amines produces molecular fluorophores (from ref. 32). .............................................................................................................................. 9 Figure 1.6. (a) some examples of interactions responsible for the CEE effect (from ref. 43). (b) a fluorescent crosslinked copolymer from polyethyleneimine and polylactic acid (from ref. 40). .................................................................................................................................................. 10 Figure 1.7. (a) excitation dependent PL from non-conjugated crosslinked PEI nanoparticles (from ref. 44), (b) excitation independent PL from non-conjugated polyamide nanoparticles (from ref. 48). ............................................................................................................................ 11 Figure 1.8. (a) polymerization and carbonization step for citric acid-based CDs. (b) Schematic representation of the emission characteristics of three photoactive species produced from the thermal treatment of mixture of citric acid and ethanolamine. During pyrolysis, the organic fluorophores (blue groups) are consumed for the buildup of the carbonized core (black sphere) so that the PL component that corresponds to the carbonized core (black bars) increases at the expenses of the component that arises from the organic fluorophores (blue bars) (from ref. 26). ................................................................................................................... 12 Figure 1.9. (left) TEM image and (right) plausible aggregation pattern of polymer CDs, resembling graphite lattice (from ref. 48). ................................................................................. 13 Figure 1.10. disassembled autoclave reactor. .......................................................................... 16 viii Figure 1.11. microwave synthesis of CDs for Ce3+ sensing (from ref. 75). ............................. 16 Figure 1.12. CDs synthesis by branched polyethyleneimine crosslinking. ............................. 17 Figure 1.13. a) overlap between CDs emission and TNP absorbance, b) CDs emission at different concentrations of TNP, c) Stern-Volmer plots for different nitroaromatics (figures from ref. 79). .............................................................................................................................. 19 Figure 1.14. FRET between CDs and the complex (Co(cys)32+). ........................................... 19 Figure 1.15. sensing mechanism of phytic acid based on the PET quenching of CDs by Fe3+ (from ref. 60). ............................................................................................................................ 20 Figure 1.16. illustration of the inner filter effect: while in a) the whole emitted light can reach the detector, in b) an additional specie is able to absorb it. The result is the decrease of the output emission. ....................................................................................................................... 21 Figure 1.17. mechanism of the amide bond formation catalyzed by EDC/NHS. Carboxylic acid reacts with 1-Ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride (EDC), forming an unstable intermediate (1). The addition of N-Hydroxysuccinimide (NHS) cause the cleavage of the EDC bond and a less labile NHS ester is formed (2). Finally an amine is added (3) and the amide bond is achieved. .............................................................................. 24 Figure 1.18. fluorescence quenching and recovery of cyclam-functionalized CDs, respectively by Cu2+ and S2ions (vials pictures from ref. 80). ..................................................................... 25 Figure 1.19. sensing mechanism based on the host-guest interaction of p-nitrophenol and cholesterol with β-cyclodextrin-functionalized CDs (from ref. 106). ........................................ 26 Figure 1.20. Mercaptosuccinic acid in acid conditions is used for the CDs functionalization. The as-produced nanoparticles are selectively quenched by Ag+ ions (ref. 107). ..................... 27 Figure 1.21. a) sulfonamide bond formation and cleavage on CDs from ref. 108 and fluorescence dependence. b) selectivity towards selenocysteine (Sec). ................................... 27 Figure 1.22. illustration of the CDs/CdS heterojunction and its photocatalytic activity towards p-nitrobenzene (from ref. 110). .................................................................................................. 29 Figure 1.23. photovoltaic device in which the CDs layer acts as spectral converter (from ref. 114). ............................................................................................................................................ 30 Figure 2.1. Photographs of CDs1, CDs2 and CDs3 in solid and in water solution (0.5 mg/mL), with and without UV irradiation. .............................................................................. 44 Figure 2.2. AFM images of CDs1 (a), CDs2 (b), CDs3 (c) and their respective height distribution (d) in black, blue and red, respectively. The average height is found at around 1 nm for all the samples. ............................................................................................................. 45 ix Figure 2.3. DLS size distribution of CDs1 (black), CDs2 (blue), CDs3 (red). All the samples show a diameter of about 1 nm. ............................................................................................... 45 Figure 2.4. DOSY spectra of CDs1, CDs2, and CDs3. ........................................................... 46 Figure 2.5. (a) UV/Vis, (b) Excitation and emission spectra of CDs1 (black), CDs2 (blue) and CDs3 (red). (c) Emission in CDs1-3 at different excitation wavelengths. ............................... 47 Figure 2.6. PL decay of CDs1 (black), CDs2 (blue) and CDs3 (red). (b) Different concentrations of CDs1 (black), CDs2 (blue), CDs3 (red) and quinine sulfate (cyan), plotted by integrated PL intensity vs. absorbance and fitted for calculating the quantum yield, expressed in %. ......................................................................................................................... 48 Figure 2.7. (a) absorbance, (c) emission (λex=370 nm) and (e) fluorescence intensity of CDs1 at different pH. (b) absorbance, (d) emission (λex=370 nm) and (f) fluorescence intensity of CDs1 switching repeatedly the pH from basic to acidic conditions and vice-versa. ............... 49 Figure 2.8. Mole percentages of C, H, N, O in CDs1, CDs2 and CDs3, and the calculated percentages for the repetitive unit of the corresponding polymeric condensation product. ..... 50 Figure 2.9. (a) IR spectra of CDs1 (black), CDs2 (blue) and CDs3 (red). .............................. 51 Figure 2.10. (a) XPS survey of CDs1 (black), CDs2 (blue) and CDs3 (red). (b) C1s, O1s and N1s spectra of CDs1 (left), CDs2 (center), CDs3 (right). ........................................................ 52 Figure 2.11. (a) 1H NMR. (b) APT 13C NMR, (c) 1H-13C HSQC and (d) 1H-13C HMBC spectra of CDs3. (e) one of the possible chain isomers of the CDs1 repetitive unit, with C and H assignation. ........................................................................................................................... 53 Figure 2.12. (a) 1H NMR. (b) APT 13C NMR, (c) 1H-13C HSQC and (d) 1H-13C HMBC spectra of CDs3. (e) one of the possible chain isomers of the CDs2 repetitive unit, with C and H assignation. ............................................................................................................................... 54 Figure 2.13. (a) 1H NMR. (b) APT 13C NMR, (c) 1H-13C HSQC and (d) 1H-13C HMBC spectra of CDs3. (e) one of the possible chain isomers of the CDs3 repetitive unit, with C and H assignation. ........................................................................................................................... 55 Figure 2.14. Optimized molecular structures of (a) two dimer (n = 2) chains and (b) one decamer (n =10) chain. (c) HOMO and (d) LUMO molecular orbitals involved in the fluorescence phenomenon. ....................................................................................................... 57 Figure 2.15. (a) Relative fluorescence intensity of CDs3 where I and Io are the intensities in presence and absence of the metal ions. (b) Emission spectra of a CDs3 water solution in the presence of 1mM concentration of Mg2+, Ca2+, Fe3+, Pb2+, Ni2+, Cu2+, Ag2+, Zn2+, Co2+, Hg2+. .................................................................................................................................................. 60 x Figure 2.16. Illustration of the CDs formation and of the photo-induced charge transfer phenomenon. ............................................................................................................................ 61 Figure 3.1. a) absorption spectra of CDs 1a-d. b) emission of CDs 1a-d for different excitation wavelengths. ............................................................................................................ 71 Figure 4.1. AFM picture and height profile of CDs 4. ............................................................ 83 Figure 4.2. DLS size distribution of CDs 4. ............................................................................ 84 Figure 4.3. ATR-IR spectrum of CDs 4. ................................................................................. 84 Figure 4.4. H1 NMR of CDs 4. ................................................................................................ 85 Figure 4.5. UV/Vis. (left) and emission spectra (right, λex=370 nm) of CDs 4. ..................... 85 Figure 4.6. ATR-IR spectra for 1,2-dithiolane derivative 1 (red) and MoS2-based materials 2 (black) and 3 (blue). ................................................................................................................. 86 Figure 4.7. Normalized Raman spectra for exfoliated MoS2 (black) and MoS2-based materials 2 (gray) and 3 (blue), obtained upon 514 nm excitation. ......................................................... 87 Figure 4.8. Thermographs for MoS2-based material 3. ........................................................... 88 Figure 4.9. SEM images for MoS2-based material 3. .............................................................. 88 Figure 4.10. UV-Vis absorption spectra of CDs– 4 upon incremental additions of (a) ammonium modified MoS2-based material 3. Inset: Enlargement of the 300-320 nm region where the isosbestic point is developed, (b) MoS2-based material 2. ...................................... 90 Figure 4.11. Photoluminescence titration assays of CDs– 4 (20 μg/mL) upon incremental additions of (a) positively charged MoS2-based material 3, and (b) neutral MoS2-based material 2. Measurements were conducted in water for samples possessing equal absorbance at the excitation wavelength of 370 nm. .................................................................................. 91 Figure 4.12. Stern-Volmer plot of CDs 4 I0/I upon incremental additions of (a) positively charged MoS2-based material 3, and (b) neutral MoS2-based material 2. ............................... 91 Figure 4.13. (a) Decay profiles for CDs– 4 upon incremental additions of positively charged MoS2-based material 3 or (b) neutral MoS2-based material 2. ................................................ 92 Figure 4.14. (a) Linear sweep voltammograms for the HER of CDs/MoS2 (black), individual CDs– 4 (red) and bare glassy carbon electrode (dotted). Inset: enlarged region near the onset. (b) Tafel plots for CDs/MoS2 (black) and individual CDs– 4 (red) showing overpotential vs current density. ......................................................................................................................... 94 Figure 5.1. H1 NMR spectra of CDs (blue) and f-CDs (red). ................................................ 103 Figure 5.2. ATR-IR spectra of as-produced CDs (blue) and 1,2-dithiolane modified f-CDs (red). ....................................................................................................................................... 103 xi Figure 5.3. UV-Vis (left) and emission (right, λex=370 nm) spectra of as-produced CDs (blue) and 1,2-dithiolane modified f-CDs (red), obtained in methanol. ........................................... 104 Figure 5.4. ATR-IR spectra of of CD-MoS2 (black) and CD-WS2 (grey). ........................... 106 Figure 5.5. Raman spectra normalized at A1g mode for (a) exfoliated MoS2 (blue) and CDMoS2 (black) at λexc 633 nm, and (b) exfoliated WS2 (blue) and CD-WS2 (grey) at λexc 514 nm. ................................................................................................................................................ 107 Figure 5.6. Raman spectra (1064 nm) for CDs (red), CD-MoS2 (black) and CD-WS2 (grey). ................................................................................................................................................ 107 Figure 5.7. Thermographs for CDs (red), exfoliated MoS2 (dotted black), exfoliated WS2 (dotted gray), CD-MoS2 (black), and CD-WS2 (grey). .......................................................... 108 Figure 5.8. Representative low-magnification HR-STEM-ADF images for CD-MoS2 (left) and CD-WS2 (right). ............................................................................................................... 109 Figure 5.9. Representative HRSTEM-ADF images for (a, d) CD-MoS2. (b) EDS acquired on the squared white area in (a). In the red regions of (c) spectra images of SR-EELS were recorded. (d) Carbon elemental map extracted from the integrated intensity of the C-K edge of the EELS spectrum image recorded in the red area in (c). (e) Three spectra from the sum of nine (3 × 3) EEL spectra extracted from the marked areas of the EELS SPIM of (c). The C-K edge (∼284 eV) is observed in (ii) and (iii) superposed with the Mo-M edge. The S-L2,3 and Mo-M edge of MoS2 are visible in the three spectra ((i)−(iii)). ............................................ 110 Figure 5.10. Representative HRSTEM-ADF images for (a, d) CD-WS2. (b) EDS acquired on the squared white area in (a). In the red regions of (c) spectra images of SR-EELS were recorded. (d) Carbon elemental maps extracted from the integrated intensity of the C-K edge of the two EELS spectra image recorded in the red area in (c). (e) Three spectra from the sum of 16 (4 × 4) EEL spectra extracted from the EELS SPIM of (c), showing the S-L2,3 and C-K (in this case only in (ii) and (iii)) edges. The C-K edge (∼284 eV) is observed in (ii) and (iii). The S-L2,3 edge is visible in the three spectra ((i)−(iii)). ...................................................... 111 Figure 5.11. (a) Absorption and (b) emission spectra (λex = 370 nm) of CD-MoS2 (black), CD-WS2 (gray), and f-CDs (red), in DMF. ........................................................................... 112 Figure 5.12. Cyclic voltammograms of (a) exfoliated MoS2, (b) exfoliated WS2, (c) CDMoS2, and (d) CD-WS2 in DMF containing 0.1 M of n-Bu4NClO4 as electrolyte. Scan rate = 100 mV/s. ............................................................................................................................... 113 Figure 5.13. Spectral changes observed during (a) first oxidation and (b) first reduction of exfoliated MoS2, and (c) first oxidation and (d) first reduction of exfoliated WS2 in DMF containing 0.2 M n-BuN4ClO4 as electrolyte. ........................................................................ 114 xii Figure 5.14. Femtosecond transient absorption spectra at the indicated delay times of exfoliated (a, c) MoS2 and (b, d) WS2, in DMF at the excitation wavelength of 425 nm (a,b) and 370 nm (c,d). The right-hand panel shows intensity-wavelength maps. ......................... 115 Figure 5.15. Femtosecond transient spectra of (a) f-CDs, (b) CD-MoS2, and (c) CD-WS2 in DMF (λexc 370 nm). The right-hand panel shows intensity-wavelength maps. ..................... 117 Figure 5.16. Femtosecond transient spectra of (a) CD-MoS2, and (b) CD-WS2, in DMF (λexc 425 nm). The right-hand panels show (ii) intensity-wavelength map and (iii) an overlap time profile of the 688 nm of CD-MoS2 (blue) and exfoliated MoS2 (red) and 652 nm of CD-WS2 (blue) and exfoliated WS2 (red). ............................................................................................ 118 Figure 5.17. Decay associated spectra of (a) exfoliated MoS2, and (b) CND-MoS2 for the transient data shown in Figure 5.14a and Figure 5.16a. ......................................................... 119 Figure 7.1. Absorption (λ 350 nm) vs Integrated emission plot for different concentration of quinine sulfate. ....................................................................................................................... 133 Figure 9.1. XPS spectra of CDs1 samples obtained at different reaction times: CDsA (green), CDsB (brown), CDsC (black). The profile is the same for all the samples. .......................... 143 Figure 9.2. Pictures taken at different times of the CDs 1b synthesis, with the UV lamp off (top) and on (bottom): a) after the addition of EDC, b) after 5 minutes, c) after 20 minutes, d) after 30 minutes and addition of NaOH solution. .................................................................. 172 Figure 9.3. AFM pictures of CDs 1a-d. In all the samples, the height of the nanoparticles is comprised between 1-2 nm. ................................................................................................... 173 Figure 9.4. DOSY spectra of CDs 1a-d. ................................................................................ 174 Figure 9.5. Visual comparison between the measured elemental composition of CDs 1a-d and the calculated elemental composition of the expected polymer structure. ............................. 175 Figure 9.6. IR spectrum of CDs 1a. 3400-2800 cm-1: O-H and N-H stretching, 1704 cm-1: C=O stretching (carboxylic acid), 1650 and 1583 cm-1: C=O stretching (amide), 1440-1350 cm-1: C-O and C-N stretching. ............................................................................................... 176 Figure 9.7. IR spectrum of CDs 1b. 3400-2800 cm-1: O-H and N-H stretching, 1704 cm-1: C=O stretching (carboxylic acid), 1650 and 1567 cm-1: C=O stretching (amide), 1440-1350 cm-1: C-O and C-N stretching. ............................................................................................... 176 Figure 9.8. IR spectrum of CDs 1c. 3400-2800 cm-1: O-H and N-H stretching, 1702 cm-1: C=O stretching (carboxylic acid), 1650 and 1582 cm-1: C=O stretching (amide), 1440-1350 cm-1: C-O and C-N stretching. ............................................................................................... 177 xiii Figure 9.9. IR spectrum of CDs 1d. 3400-2800 cm-1: O-H and N-H stretching, 1708 cm-1: C=O stretching (carboxylic acid), 1650 and 1588 cm-1: C=O stretching (amide), 1440-1350 cm-1: C-O and C-N stretching. ............................................................................................... 177 Figure 9.10. APT 13C NMR spectrum of CDs 1a. Carboxylic acid and amide C: 182-171 ppm, quaternary C: 75-72 ppm, methylene C: 45-36 ppm. The positive signal at 21 ppm is due to DIC impurities. ................................................................................................................... 178 Figure 9.11. 1H NMR spectrum of CDs 1a. methylene H: 3.9-2.3 ppm. The signal at 1.2 ppm is due to DIC impurities. ........................................................................................................ 178 Figure 9.12. 1H-13C HSQC spectrum of CDs 1a. .................................................................. 179 Figure 9.13. 1H-13C HMBC spectrum of CDs 1a. ................................................................. 179 Figure 9.14. APT 13C NMR spectrum of CDs 1b. Carboxylic acid and amide C: 180-170 ppm, quaternary C: 75-72 ppm, methylene C: 46-35 ppm. The signals at 160, 55, 42, 24, 14 ppm are due to EDC impurities. ............................................................................................. 180 Figure 9.15. 1H NMR spectrum of CDs 1b. methylene H: 3.9-2.3 ppm. The signals at 3.0, 2.8, 1.8, 1.0 ppm are due to EDC impurities. ................................................................................ 180 Figure 9.16. 1H-13C HSQC spectrum of CDs 1b. .................................................................. 181 Figure 9.17. 1H-13C HMBC spectrum of CDs 1b. ................................................................. 181 Figure 9.18. APT 13C NMR spectrum of CDs 1c. Carboxylic acid and amide C: 180-172 ppm, quaternary C: 76-72 ppm, methylene C: 46-35 ppm. The signals at 161, 39, ppm are due to DMF traces. ........................................................................................................................ 182 Figure 9.19. 1H NMR spectrum of CDs 1b. methylene H: 3.8-2.3 ppm. The signals at 2.8 ppm is due to DMF traces. ..................................................................................................... 182 Figure 9.20. 1H-13C HSQC spectrum of CDs 1c. .................................................................. 183 Figure 9.21. 1H-13C HMBC spectrum of CDs 1c. ................................................................. 183 Figure 9.22. APT 13C NMR spectrum of CDs 1d. Carboxylic acid and amide C: 181-172 ppm, quaternary C: 75-72 ppm, methylene C: 46-35 ppm. The signals at 65, 29, 25 ppm are due to the THF and its cleavage product by means of HCl. ................................................... 184 Figure 9.23. 1H NMR spectrum of CDs 1d. methylene H: 3.8-2.3 ppm. The signals at 4.0, 1.7 ppm are due to the THF and its cleavage product by means of HCl. ..................................... 184 Figure 9.24. 1H-13C HSQC spectrum of CDs 1d. .................................................................. 185 Figure 9.25. 1H-13C HMBC spectrum of CDs 1d. ................................................................. 185 Figure 9.26. UV/vis spectra of CDs 1a-d. ............................................................................. 186 Figure 9.27. (left) excitation spectra for different emissions and (right) emission spectra for different excitations of CDs 1a. .............................................................................................. 186 xx List of Schemes Scheme 2.1. (Top) Reaction of EDA with CA through two synthetic pathways to form CDs1 and CDs3. (Bottom) Reaction of EDA with TA to form CDs2. .............................................. 43 Scheme 3.1. Polycondensation of CA and EDA through four synthetic pathways, for the obtaining of CDs 1a-d. ............................................................................................................. 69 Scheme 3.2. Synthesis of CDs 2a-f. ......................................................................................... 72 Scheme 3.3. a) polycondensation of CA and EDA mediated by coupling agent. b) the addition of a primary amine consumes the activated carboxylic acids and stops the polymerization. .. 74 Scheme 4.1. Preparation of CDs4. ......................................................................................... 82 Scheme 4.2. Functionalization of MoS2 leading to ammonium modified MoS2-based material 3. ............................................................................................................................................... 83 Scheme 4.3. Preparation of the CDs/MoS2 electrostatic complex. .......................................... 89 Scheme 5.1. synthesis of CDs and functionalization with lipoic acid. .................................. 101 Scheme 5.2. Illustrative preparation of CD-MoS2 and CD-WS2 upon covalent 1,2-dithiolane functionalization of exfoliated semiconducting MoS2 and WS2 nanosheets. ........................ 105 Scheme 11.1. Synthesis of raw CDs. ..................................................................................... 241 xxi List of Tables Table 2.1. Measured diffusion coefficient (D), calculated hydrodynamic radius (rH) and diameter of CDs1, CDs2, and CDs3. All samples show a diameter between 1.2-1.6 nm. ...... 46 Table 2.2. Elemental analysis of CDs1, CDs2 and CDs3 and the calculated percentage (mol%) for the repetitive unit of the corresponding polymeric condensation products. .......... 50 Table 2.3. Absorption (Eabs) and emission (Eem) energies ....................................................... 58 Table 3.1. amines employed for the polycondensation, emission maximum, QY and pictures (in water, UV light off and on) of CDs 2a-f. ............................................................................ 72 Table 3.2. amine employed for the functionalization, emission maximum, QY and pictures (in water, UV light off and on) of CDs 3a-e. ................................................................................. 74 Table 8.1. Measured diffusion coefficient (D) and the calculated hydrodynamic radius (rH) and diameter of CDs1 samples obtained at different reaction times: CDsA, CDsB, CDsC. . 137 Table 8.2. Average molecular weight in number (Mn ), in weight (Mw ) and polydispersity (Ɖ) of CDs1 samples obtained at different reaction times: CDsA, CDsB and CDsC. ........... 138 Table 8.3. Elemental Analysis of CDs1 samples obtained at different reaction times: CDsA, CDsB, and CDsC. ................................................................................................................... 141 Table 9.1. The measured diffusion coefficients (D) and the calculated hydrodynamic rays (rH) and diameters of CDs 1a-d. Assuming a globular shape, the calculated size of the nanoparticles is comprised between 1.2-2.0 nm..................................................................... 174 Table 9.2. C, H, N, O and S mole percentages of CDs 1a-d, obtained by elemental analysis. Additionally, the calculated elemental composition of the expected polymer repetitive unit is reported for comparison. ........................................................................................................ 175 Table 9.3. The measured diffusion coefficients (D) and the calculated hydrodynamic rays (rH) and diameters of CDs 2a-f. Assuming a globular shape, the calculated size of the nanoparticles is comprised between 1.2-2.0 nm, with the exception of CDs 2c, which show a size of 4.8 nm. ........................................................................................................................ 191 Table 9.4. C, H, N, O and S mole percentages of CDs 2a, obtained by elemental analysis. Additionally, the calculated elemental composition of the expected polymer repetitive unit is reported for comparison. ........................................................................................................ 191 Table 9.5. C, H, N, O and S mole percentages of CDs 2b, obtained by elemental analysis. Additionally, the calculated elemental composition of the expected polymer repetitive unit is reported for comparison. ........................................................................................................ 192 xxii Table 9.6. C, H, N, O and S mole percentages of CDs 2c, obtained by elemental analysis. Additionally, the calculated elemental composition of the expected polymer repetitive unit is reported for comparison. ........................................................................................................ 192 Table 9.7. C, H, N, O and S mole percentages of CDs 2d-f, obtained by elemental analysis. Additionally, the calculated elemental composition of the expected polymer repetitive unit is reported for comparison. ........................................................................................................ 193 Table 9.8. The measured diffusion coefficients (D) and the calculated hydrodynamic rays (rH) and diameters of CDs 3a-e. Assuming a globular shape, the calculated size of the nanoparticles is comprised between 1.2-2.4 nm..................................................................... 216 Table 9.9. C, H, N, O and S mole percentages of CDs 3a-e, obtained by elemental analysis. ................................................................................................................................................ 216 Table 11.1. C, H, O and N composition of fraction 1-4. ....................................................... 244 xxiii List of abbreviations and acronyms 2D: Two dimensional AcOEt: Ethyl Acetate AFM: Atomic Force Microscopy APT: Attached Proton Test BOC: tert-Butoxycarbonyl CA: Citric acid CDs: Carbon dots CEE: Crosslink-Enhanced Emission CV: Cyclic voltammetry DCM: Dichloromethane DFT: Density Functional Theory DIC: N,N′-Diisopropylcarbodiimide DLS: Dynamic light scattering DMF: Dimethylformamide DOSY: Diffusion ordered spectroscopy EDA: Ethylenediamine EDC: N-(3-Dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride EDS: Energy-dispersive X-ray spectroscopy EELS: Electron Energy Loss Spectroscopy FTIR: Fourier-transform infrared spectroscopy HB: Hydrogen Bond HMBC: Heteronuclear Multiple Bond Correlation HOMO: Highest occupied molecular orbital HAADF-STEM: high-angle annular dark-field scanning transmission electron microscopy HRTEM: High resolution transmission electron microscopy HSQC: heteronuclear Single Quantum Correlation LUMO: Lowest unoccupied molecular orbital MeOH: Methanol xxiv MWCO: Molecular weight cut-off NHS: N-hydroxysuccinimide NMR: Nuclear magnetic resonance PTFE: Polytetrafluoroethylene QY: Quantum Yield SEC: Size exclusion chromatography TA: Tricarballylic acid TDDFT: Time Dependent Density Functional Theory TEM: Transmission electron microscopy TMDs: Transition metal dichalcogenides THF: Tetrahydrofuran TGA: Thermogravimetric analysis UV/Vis: Ultraviolet/visible spectroscopy XPS: X-ray photoelectron spectroscopy xxv ABSTRACT Carbon dots (CDs) are an emerging class of organic nano-sized particles, whose exceptional photoluminescence surely is one of the most intriguing properties. Thanks to their photoinduced energy transfer and charge transfer ability, CDs are an extremely valuable material for the preparation of photocatalysts and donor-acceptor composites for energy conversion applications. Moreover, the presence of metal ions as well as organic species affects heavily the CDs’ photoluminescence intensity. For this reason, CDs are widely employed for the design of highly sensitive sensors and biosensors. Finally, CDs demonstrated to be a safe and biocompatible material, perfectly suitable for medical applications such as imaging and drug delivery. Despite their consistent use for various applications, the chemical nature of CDs and its relationship with their outstanding optical properties is still a topic under debate, being the actual knowledge incomplete and at times contradictory. In fact, the wide variety of precursors and synthetic methods employed for the preparation of CDs is reflected in the high variability of their structures, hampering the identification of the common fundamental elements behind the fluorescence emission. Additionally, the lack of understanding and control on the synthetic process undermines the ability to design and tailor the CDs structure, which instead is a key point for improving their properties or making them suitable for further applications. In this thesis, the first part is focused on fundamental structural and optical studies on citric acid-based CDs, which, jointly with theoretical calculations, provided new insights into the CDs chemical nature and furnished a general explanation for their blue fluorescence emission. Next, these findings were exploited for the development of a novel room-temperature and versatile synthesis method, which allows the rational design of the CDs polymer structure and at the same time their in-situ functionalization, granting in this way the full control on the final chemical structure. In the last part of the thesis, the preparation of CDs-based material and their chemical and optical characterization is reported. In here non-covalent and covalent approaches were both employed for combining CDs with transition metal dicalchogenides (TMDs) and creating photo-active donor-acceptor materials for energy applications. xxvi Chapter 1 provides a general overview regarding CDs and CDs based materials. It includes a chronological review on the fundamental studies investigating the CDs chemical nature from their discovery until the actual state of the art, as well as a presentation of the parallel development of theories about the origin of the fluorescence. Next, the most common methods of synthesis and functionalization of CDs are presented. Finally, together with many examples, the employment of CDs in the field of sensing, medicine, catalysis and energy conversion is described. Chapter 2 consists of an accurate structural and optical investigation of polymeric CDs. In here, the choice of the precursors and synthetic methods employed was determinant for identifying the structural features responsible for the bright blue fluorescence emission. By sample comparison, it was proved that the polycondensation between citric acid and ethylenediamine into a polyamide is the only necessary condition for obtaining the excitationindependent blue emission. In order to confirm these results and understand the role of the organic moieties involved in the fluorescence, DFT calculations based on the experimental results were performed. The computational models showed that strong intra-molecular Hbonds are responsible for the high conformational rigidity of CDs, which hampers vibrations and rotations, promoting thus the radiative relaxation. Additionally, amide and carboxylic acids were identified respectively as HOMO and LUMO of the fluorescent process, which consists of an intra-molecular charge transfer. Chapter 3 presents a novel, versatile pathway towards the rational design of a wide variety of polymeric fluorescent CDs with well-defined structures. The exploited reaction consists of the room-temperature carbodiimide-mediated condensation between citric acid and amines. Moreover, this method allows the in-situ introduction of several types of desired moietes onto the surface of the CDs, thus achieving the CDs synthesis and functionalization in only one step. Chapter 4 focuses on the preparation, characterization, and study of the photophysical and electrocatalytical properties of CDs/MoS2 electrostatic complexes. Negatively charged CDs prepared from microwave irradiation of citric acid and ethylenediamine were employed in the titration essay of a dispersion of positively charged MoS2 oligolayers. The optical properties of the forming electrostatic complex were studied ongoing. Efficient fluorescence quenching of CDs by MoS2 was observed and ascribed to photoinduced electron and/or energy transfer as the decay mechanism for the transduction of the singlet excited state of CDs. Finally, the xxvii electrocatalytic performance of CDs/MoS2 was assessed towards the hydrogen evolution reaction and found superior as compared to that owed to the individual CDs species. Chapter 5 covers the preparation of covalent CDs-TMDs (MoS2 and WS2) hybrids and their characterization, including spectroscopic, thermal and electron microscopy imaging methods. The preparation of the material was started by functionalizing CDs with disulfide moietes. Thus, the high affinity of disulfide groups towards Mo and W was exploited for attaching CDs onto the metal atoms defects located at the exfoliated TMDs edges. Steady-state and time-resolved fluorescence spectroscopy determined the occurrence of fast energy and/or charge transfer processes between CDs and TMDs. Furthermore, transient absorption spectroscopy studies revealed that upon MoS2 photoexcitation charge transfer from an exciton dissociation path of MoS2 to CDs, within CD-MoS2, was observed. In contrast, CD-WS2 did not display such behavior due to energetic reasons. The electronic processes taking place in this novel material certainly can be of interest for the development of donor-acceptor components in view of energy conversion applications. Finally, Chapter 6 provides the general conclusion of the findings and an outlook in what concerns possible opportunities and future research work. xxviii RESUMEN Los Puntos de Carbono, en inglés Carbon Dots (CDs), son una clase de partículas orgánicas nanométricas, cuya excepcional fluorescencia es ciertamente una de sus propiedades más fascinantes. Gracias a propiedades como la transferencia de energía y/o de carga, los CDs resultan ser materiales muy atractivos para la preparación de foto-catalizadores y compuestos donor-aceptor para aplicaciones en el campo de la conversión de energía. Además, la presencia de iones metálicos así como de especies orgánicas puede afectar considerablemente a la intensidad de la fluorescencia de los CDs. Por esta razón, los CDs se utilizan ampliamente para el desarrollo de sensores y biosensores altamente sensibles. Finalmente, los CDs son materiales seguros y biocompatibles, perfectamente adecuados para aplicaciones médicas como agentes de contraste o vectores de fármacos. A pesar de su amplio uso en varias aplicaciones, la naturaleza química de los CDs y la relación con sus increíbles propiedades ópticas son todavía argumento de debate, siendo el conocimiento actual incompleto y a veces contradictorio. Esto se debe también a la gran variedad de precursores y métodos de síntesis empleados en la preparación de los CDs, lo que se refleja en la dificultad de identificar los elementos fundamentales y comunes relacionados con el proceso de fluorescencia. Además, la falta de comprensión y de control sobre el proceso sintético limita la posibilidad de diseñar y modificar la estructura de los CDs, lo que es un factor clave para mejorar sus propiedades o adaptarlas para aplicaciones específicas. La primera parte de este trabajo de tesis se centra en estudios fundamentales de la estructura y propiedades ópticas de los CDs derivados de ácido cítrico y etilendiamina. Resultados experimentales de esta investigación junto a cálculos teóricos han aportado nuevo conocimiento con respecto a la naturaleza química de los CDs y también han contribuido a dar una explicación general a su emisión de fluorescencia azul. Posteriormente, se han aprovechado estos descubrimientos para el desarrollo de un nuevo y versátil método de síntesis, que consigue el diseño de la estructura polimérica de los CDs y, al mismo tiempo, su funcionalización in-situ, asegurando de esta manera un control completo sobre la estructura química final. En la última parte de la tesis, se reporta la preparación de materiales basados en los CDs y su caracterización química y óptica. Aquí se han utilizado estrategias de enlace covalente y non-covalente se han utilizado para combinar los CDs con dicalcogenuros monocapa de metales de transición (TMDs: transition metal dichalcogenides) y crear materiales donor-aceptor fotoactivos para aplicaciones energéticas. xxix El Capítulo 1 consiste en un resumen general en relación a los CDs y a sus materiales derivados. Incluye un repaso cronológico de los estudios fundamentales que investigaron la naturaleza química de los CDs desde su descubrimiento hasta el momento actual, así como una descripción del desarrollo paralelo de teorías sobre el origen de la fluorescencia. El Capítulo 2 presenta una rigurosa investigación estructural y óptica de los CDs. Aquí, la elección de los precursores y de los métodos sintéticos ha sido determinante para identificar los componentes estructurales responsables de la intensa emisión de fluorescencia azul. Comparando los resultados de las diferentes muestras, se ha establecido que la policondensación entre ácido cítrico y etilendiamina es la única reacción necesaria para la obtención de fluorescencia independiente de la excitación. Para confirmar estos resultados y para entender el papel de los grupos organicos involucrados en la fluorescencia, se han realizado cálculos de DFT (teoría funcional de la densidad) basados en los datos experimentales. Los modelos computacionales enseñan que hay fuertes enlaces de hidrogeno intra-moleculares responsables de la alta rigidez conformacional de los CDs, lo que limita la vibración y rotación de enlaces implicados, fomentando así la relajación radiativa. Además, amidas y acidos carboxílicos se han identificado respectivamente como HOMO y LUMO del proceso fluorescente, que consiste en una transferencia de carga intra-molecular. El Capítulo 3 presenta un camino nuevo y versátil hacia la síntesis racional de una amplia variedad de CDs fluorescentes con estructuras poliméricas bien definidas. La reacción utilizada consiste en la condensación a temperatura ambiente mediada por carbodiimida entre ácido cítrico y etilenodiamina. Además, este método posibilita la introducción in-situ de distintos tipos de grupos funcionales en la superficie de los CDs, logrando así síntesis y funcionalización en un solo paso. El Capítulo 4 se enfoca en la preparación, caracterización y estudio de las propiedades fotofísicas y electro-catalíticas de complejos electroestáticos de CDs/MoS2. CDs cargados negativamente, preparados por irradiación en microondas de ácido cítrico y etilendiamina, se han empleado en el ensayo de valoración de una dispersión de oligo-capas de MoS2 cargadas positivamente. Asimismo, se han estudiado las propiedades ópticas del compuesto electroestático durante la formación. Se ha observado una eficiente desactivación (quenching) de la fluorescencia de los CDs por parte del MoS2, lo que se ha atribuido a un mecanismo de decaimento desde el estado excitado de singlete de los CDs, consistente en una trasferencia de 6 In these pioneering works, some common points should be highlighted. The carbon source employed here consisted of pure carbon materials, such as graphite, carbon nanotubes and active carbon. The synthetic procedures involved the destruction and oxidation of the material, aiming to create the defects acting as surface traps. Additionally the fluorescence emission of the as-produced nanoparticles commonly showed an excitation-dependent behavior. Considering that these CDs are expected to have large sp2 conjugated domains delimited by oxygenated groups, the excitation-dependent PL was attributed to the variety of trap states on the particles surface, as well as to the quantum size effect, i.e. the radiative recombination of excitons whose energy depends from the size of the conjugated domain. With these premises, passivation was considered to play a role in the stabilization of the energy traps, enhancing the emission. In 2008, the group of P. Giannelis et al. reported an entirely different approach for the synthesis of photoluminescent organic nanoparticles17-18. Citrate ammonium salts and other molecular precursors were heated at 300 °C by hydrothermal treatment or pyrolysis, obtaining nanoparticles of <10 nm size and quantum yield (QY) of 3%. Their optical properties showed many similarities with the ones of the CDs obtained from fragmentation and oxidation of carbon materials and an analogy with them was made. Thus, it was suggested that the carbonization of the molecular precursor leads to the formation of a carbon core, composed by co-existing aromatic and aliphatic regions, similarly to graphite oxide. Additionally, TEM images, high water solubility and the presence of amide bonds proved that a polymeric corona covers the surface of the nanoparticles. Therefore, polymerization was proposed to be the reaction step preceding the core carbonization. The following studies explored more synthetic routes, finding that hydrothermal treatment1921, pyrolysis by microwave irradiation22-24 and strong acid treatment11 were all effective onestep methodologies for the obtaining of fluorescent nanoparticles, achieving better quantum yields and avoiding annoying oxidation and passivation steps with respect to the top-down approaches. Possibly because these new materials resembles in size and optical properties the firsts fluorescent carbon nanoparticles discovered, they were as well called carbon dots (or Cdots, carbon nanodots, carbon quantum dots etc.), widening and blurring the limits of this definition. This vagueness most likely was encouraged by looking for transversal principles beyond the PL, and the attention was initially focused on the carbonized core and its interaction with the amorphous surface, since the synthetic procedures were generally harsh and high temperature unavoidable forms sp2 unpredictable structures. 7 The complex fluorescence behavior of bottom-up synthetized CDs is often described as a combination of processes, covering excitons recombination (Figure 1.4a), presence of trap states which may promote the radiative decay (Figure 1.4b), and interaction of the oxygenated groups at the edges with the conjugated domains25. Nevertheless, these hypothesis imply (and the dictionary suggests) the existence of highly delocalized states, i.e. a semi-conducting band from where the electron can decay through different pathways. Under this point of view, the existence of a graphitized core, offering dlocalized ground and excited states, is the necessary condition for the appearance of PL, while oxygenated groups simply promote the radiative decay acting as localized acceptor traps for the excited electrons. Moreover in this framework nitrogen is commonly depicted as a dopant (“N-doped carbon dots” appeared more than 200 times in the article titles between 2010 and 2018; source: Scopus), suggesting its participation in the process of electron delocalization. Figure 1.4. (a) PL due to the quantum size effect. The sum of different conjugated domains with distinct energy band gaps can be responsible for the observed excitationdependent PL behavior. (b) Oxygen groups located at the edges of the conjugated domains can act as localized energy traps, which promote the radiative relaxation. 8 However, in 2012 P. Giannelis et al. studied the formation of CDs by pyrolysis of citric acid and ethanolamine at different temperatures and observed that carbonization only starts above 200 °C.26 Interestingly, CDs prepared at lower temperatures display very intense PL emission independent from the excitation (QY=50%). Increasing the temperature, this features drastically fall, almost disappearing at 300 °C, while with the growth of the carbonized core a new type of excitation-dependent PL is observed, but with considerably lower quantum yield (4% at 300 °C). Thus, it was proposed that at relatively low temperature (<200 °C) the formation of molecular fluorophores occurs. Increasing the temperature the fluorophore is consumed by the carbonization process, and the excitation-dependent PL rises at the expenses of the excitation-independent component. Further studies brought additional arguments for the existence of the molecular fluorophores27-29. The structural recognition of the molecular fluorophore is largely pursued and since 2015 several reports linked the PL to particular aromatic structures, formed when citric acid and amines precursors were employed. The first report in this direction, by B. Yang et al., recognized the origin of the molecular state fluorescence in the IPCA conjugated molecule (imidazo[1,2-a]pyridine-7-carboxylic acid), which was found attached to the carbon core and whose optical properties are in good agreement with the CDs excitation-independent blue PL (Figure 1.5a,b).30-31 Naturally, the fluorophore structure depends on the reagents employed (Figure 1.5c,d). Many publications identified similar pyridine-like/carboxylic acid structures as origin of the molecular state PL32-36, also underlining the influence of supramolecular Hbond or aggregation. 9 Figure 1.5. reactions that occur during the hydrothermal treatment of citric acid and ethylenediamine towards the formation of molecular fluorophores, polymer clusters and the carbonized core (from ref. 30) (b) IPCA fluorophore and its optical properties (from ref. 30). (c) molecular fluorophore and core contributions to the absorption of CDs (from ref 36). (d) the hydrothermal treatment of citric acid with different amines produces molecular fluorophores (from ref. 32). Nevertheless, in the last few years another relevant type of PL process was discussed in relation to the CDs optical behavior. It was established that non-conjugated polymers are capable of displaying bright PL emission in condition of high rigidity. Aggregation37-38, hyperbranching39-40, self-assembly41 or crosslinking40, 42 are some of the conditions that enable the PL emission in non-conjugated polymers (Figure 1.6). This phenomenon was therefore called crosslink enhanced emission (CEE) effect. B. Yang documented strong PL emission in various CDs systems prepared in mild conditions, regardless the absence of carbon core or simple C=C double bonds and thus excluding also the presence of aromatic molecular fluorophores2, 43-46. The PL of CDs was therefore related to the immobilization of the so-called sub-fluorophores located on the polymer structure, which are organic chromophores (such as C=O, C=N, N=O, corresponding for example to amide and carboxylic acid groups) that, due to the crosslinked and rigid structure of the CDs, become able to reemit the absorbed photons by means of radiative pathways. 10 Figure 1.6. (a) some examples of interactions responsible for the CEE effect (from ref. 43). (b) a fluorescent crosslinked copolymer from polyethyleneimine and polylactic acid (from ref. 40). The emission due to the CEE effect may be affected or not from the excitation wavelength employed (Figure 1.7). When the energy distribution of the polymer conformations originating the PL is narrow, i.e. the PL centers are similar, the emission is independent from the excitation. On the other hand, when the emitting states are originated from different conformations, emission can be tuned by excitation. In fact, the lowering of the excitation energy leaves the various higher emission energy levels inaccessible and the exclusion of their contribution from the total PL is not only reflected in the intensity decrease, but also the emission red-shift47. The rate of aggregation can also broaden the energy levels and make PL dependent from excitation37. 11 Figure 1.7. (a) excitation dependent PL from non-conjugated crosslinked PEI nanoparticles (from ref. 44), (b) excitation independent PL from non-conjugated polyamide nanoparticles (from ref. 48). 1.2.3. Discussion From the CDs discovery up to now, four coexisting interpretations of the PL phenomena were formulated in the following chronological order: 1. Quantum size effect of aromatic domains with different sizes (excitation dependent). 2. Trap states from oxygen moieties and from passivation by nitrogen containing molecules (excitation dependent or independent, according to the energy distribution of the traps). 3. Conjugated molecular fluorophores attached on the CDs surface (excitation independent). 4. CEE effect of non-conjugated polymer (excitation dependent or independent, according to the distribution of conformations and aggregation effects). Naturally, the listed PL emitters are not mutually exclusive. On the contrary, the complex optical behavior of the CDs has been frequently described as the sum of different contributions. Nevertheless, the gathered information can be exploited to discuss in hindsight the way that statements about the fluorescence-structure relationship should be argued from the experimental data. A revision of the most widespread interpretations in favour of an updated point of view that considers and includes the latest findings is highly desirable for providing better basis for the future development of the field of CDs. As aforementioned, the PL due to quantum size effect, trap states and passivation implies the presence of large conjugated domains in the CDs structure. When CDs were discovered, the employed methods generally involved the use of pure carbon material as precursor, as well as very high temperatures and harsh treatments. Thus, it is reasonable that the observed PL in 12 these nanoparticles has to be somehow related to their conjugated structures. Nowadays these materials are most likely classified as GQDs, a class of carbon nanoparticles consisting of nano-sized single or multilayer of graphene oxide. The bottom-up approaches however afford completely different results: the use of organic molecules as precursors requires diverse procedures, which generally involve two consequential synthetic steps: the formation of the polymer nanoparticles and their partial carbonization. Possibly due to the fact that the bottomup approach followed chronologically the top-down approach, carbonization has been considered for long time the key step for enabling the PL properties of CDs and it has been pursued by employing harsh procedures, with temperatures not lower than 150-250 °C. High temperatures promote dehydration and decarboxylation, which are reactions required for the formation of aromatic structures. As previously reported, these structures may involve individual conjugated molecular fluorophores, whose PL is independent from the excitation, or, at higher temperature, random graphitic domains. PL dependency from the excitation has been generally related to these last products of carbonization. However excitation dependency should not be considered a proof of the contribution of graphitic domains to the PL, since comparable excitation-dependent emission was also reported for pure non-conjugated polymer nanoparticles. On the contrary, since polymerization is the preliminary step for the carbonization (Figure 1.8a), the hypothesis that PL emission is due to the CEE effect should be considered first, assuming that the synthetic method does not achieve the complete carbonization (>300 °C) (Figure 1.8b). Figure 1.8. (a) polymerization and carbonization step for citric acid-based CDs. (b) Schematic representation of the emission characteristics of three photoactive species produced from the thermal treatment of mixture of citric acid and ethanolamine. During pyrolysis, the organic fluorophores (blue groups) are consumed for the buildup of the carbonized core (black sphere) so that the PL component that corresponds to the carbonized core (black bars) increases at the expenses of the component that arises from the organic fluorophores (blue bars) (from ref. 26). 13 Additionally, even the simple presence of graphitized domains should be proved carefully. It was reported that also polymers can form crystalline structure inside the CDs, which could be potentially mistaken as graphite lattice when analyzed by TEM and XRD (Figure 1.9).48 These techniques are employed routinely for demonstrating the CDs graphitic nature, jointly with XPS analysis. However XPS structural attribution is rarely supported by other techniques for the precise determination of the carbon structures and therefore it has a somewhat arbitrary taste. For an improved structural characterization, NMR spectroscopy can offer a wide choice of experiments that allow the recognition of the CDs features, easily revealing the rate and the type of carbonized structures, as well as improving the understanding of the polymer component. Finally, the passivation process should be discussed under the light of the CEE effect. Passivation with amine-containing molecules or polymers has been largely exploited for originating or enhancing the PL of carbon nanoparticles carrying carboxylic acids. This process, promoted by heating, has been frequently related to the stabilization of the trap states, as well as to the doping of the conjugated structures. Nevertheless, amide is known to increase the molecule structural rigidity, possibly promoting the CEE effect, and could also participate to the PL emission as subfluorophore. A role of nitrogen atoms, independent from the presence of conjugated systems, should be certainly considered when amines and carboxylic acids are found among the CDs precursors. Figure 1.9. (left) TEM image and (right) plausible aggregation pattern of polymer CDs, resembling graphite lattice (from ref. 48). 1.2.4. Conclusion In summary, a brief chronological report was intended to draw the attention of the reader on the interconnection between birth and evolution of the CDs synthesis approaches and the development of theories regarding the structure/PL relationship. Under the light of the latest 14 findings, the validity limits of some established interpretations concerning the PL chemical nature of CDs were reconsidered, pointing the attention on the underestimated role of the polymer component, which, in contrast to the carbonized core, is always present when molecular precursor are employed in the synthesis. In particular, it is shown that facts such as PL dependence from the excitation and presence of C=C double bonds cannot be presented as the sole evidences of a contribution from the carbonized component to the PL, when the presence of the polymer component could furnish alone a self-sufficient explanation. For equivalent reasons it is proposed to consider nitrogen not only for its possible electronic participation on the conjugated structures, but also and primarily as amide, when carboxylic acids and amines are employed in the CDs synthesis. Amide in fact could play a role in the PL process independently from the existence of a carbonized core. Finally, a wider use of the NMR spectroscopy is encouraged for safely ascertaining the presence and the extent of the carbonized component, as well as the nature of the polymer structure. 1.3. Precursors and synthesis methods of CDs 1.3.1. Introduction In this chapter a description of the precursors and the synthetic methods employed for the synthesis of CDs is presented. In fact, these two points are fundamental for understanding their chemical and optical properties and they also will furnish an idea of the extreme variety of structures and materials that goes by the name of CDs. 1.3.2. Precursors Fluorescent nanoparticles have been produced from many and very different precursors. Indeed, any oxygenand nitrogen-containing organic molecule can be employed, without neither the need of complex synthetic processes nor difficult purificiation steps (basically, the majority of these procedures involves the simple heating of the starting material in solution, as described in the following sections). For this reason, researchers were very imaginative at the moment of the precursors choice. For example natural substances were extensively explored as source for the obtaining of CDs and included grass49, hair fibers50, orange juice51, melon peel52, rice flour53, papaya54 and urine55 (pee-dots). Nevertheless, more ordinary reagents are usually preferred. Actually, CDs are normally obtained from simple molecules rich in oxygen groups, typically sugars, aminoacids, ascorbic acid and citric acid. Probably, citric acid is the most common reagent; so much that citrate-based CDs recently deserved 15 alone a full review.56 It is widely accepted that the presence of nitrogen is responsible for a strong enhancement of the fluorescence emission and therefore nitrogen-containing molecules are also normally found among the precursors. In particular ammonia32, urea57, ethylenediamine45 and other simple amines are often employed. Also L-lysine has been used for this purpose58-60, while L-cysteine is added when the introduction of sulfur is desired.61-62 Finally, polymers constitute another exploited category of precursor, among which polysaccharides (such as chitosan)63, polyethylene glycol64, polyacids45 and polyamines44. In general, the use of molecular precursors is more widespread with respect to natural materials and polymers, probably because it allows the obtaining of simpler structures with better defined properties. 1.3.3. Synthetic methods Due to the increasing ability to understand the relationship between structure and fluorescence of CDs, some older synthetic procedures (such as laser ablation/oxidation route, combustion, additional passivation steps) were replaced along the years with more efficient approaches (see Chapter 1.2 for details). In general, the majority of the employed methodologies involve harsh temperature conditions, in order to achieve the condensation and dehydration reactions responsible for the growth of the CDs. In this section the most nowadays common procedures for the obtaining of CDs, i.e. hydrothermal treatment, microwave irradiation, other carbonization procedures and polymerization, are described and illustrated with few examples. 1.3.3.1 Hydrothermal treatment The hydrothermal method is based on the employment of autoclave (Figure 1.10) for treating at high temperature and high pressure the water solution containing the precursors. Typically the reported synthesis temperatures for CDs are found between 150-250 °C and the time of reaction is usually comprised between 1-6 hours. This method is probably the most widespread, because it allows to obtain nanoparticles with controllable size, surface modifications and high stability.32, 51, 60-62, 64-70 B. Yang et al. exploited the hydrothermal method for synthetizing a wide variety of CDs from citric acid and different amines.27 Moreover they performed the synthesis at different temperatures, from 150 to 250 °C, observing a trend in the quantum yield, which laid the basis for the understanding of the polymer structure role in the fluorescence behavior. 22 1.5. Toxicity of CDs Since their discovery, CDs have been immediately recognized as an interesting alternative to semiconductor quantum dots (QDs) in biological and medical applications. In fact, while the employment of QDs is limited by the intrinsic toxicity associated to their metallic composition, CDs are made of biocompatible elements and therefore offer a safer choice. Several studies were performed in order to assess the toxicity of various CDs in vitro and in vivo. Y. Zhang et al. monitored the growth of green beans in a CDs solution, in a CdTe QDs solution and in water.94 While the bean sprouts resulted pathological in the QDs solution, their growth was comparable in the CDs solution and in water. The fluorescence properties of the bean plant proved that CDs are able to permeate throughout the plant cells with good biocompatibility and that they are non-toxic and do not hinder plant growth. In the work of D. Cui et al. the acute toxicity, subacute toxicity and genotoxicity of CDs was systematically tested, concluding that no significant toxicity effects were detected in mice organs.95 Other studies reported the employment of CDs for in vivo and in vitro imaging of various tumor cell lines, without finding any significant toxicity.96 Also other cell types were investigated. S. K. Kailasa used CDs as probes for imaging of bacterial and fungal cells.97 In both cases the internalization occurred successfully without any toxic effect on the cells. CDs also proved to be able to decrease the toxicity of other drug vectors. Polyethyleneimine (PEI) is an efficient vector for gene delivery, but its use is associated with cytotoxicity. In the work of W. Liu et al., PEI-passivated CDs showed a DNA transfection efficiency in vitro comparable to PEI alone, but with decreased cytotoxicity.98 Finally, a deep study on the pharmacokinetic of PEG-functionalized CDs was performed by the group of X. Chen.99 The effects of intravenous, intramuscular and subcutaneous routes were compared, observing that in the three cases CDs are rapidly concentrated in kidneys and liver and efficiently excreted from the body, without showing appreciable toxicity. In conclusion, even if CDs chemical structure and properties can be different from one another due to the different precursors and methods employed for their synthesis, a huge amount of toxicity studies in vitro and in vivo on numerous types of cells and in mice indicates that these nanoparticles are generally safe and biocompatible. Also for this reason the impact of CDs on the preparation of nanotechnologies and materials for bioimaging, biosensing, drug delivery and other medical applications is potentially high and thus this field is rapidly developing. 23 1.6. Functionalization of CDs for biomedical applications 1.6.1. Introduction Many interesting applications are perfectly suited for the employment of CDs, including metal sensing, biomolecule sensing, bioimaging and drug delivery. Indeed bright photoluminescence and high water solubility and biocompatibility makes CDs very promising candidates for analytical and medical purposes. However, in view of their use a critical requirement is that the nanoparticles must be selective, that is to say that CDs should display a strongly preferential interaction with the target to reach/analyze. Actually, the method and the precursors frequently employed for the synthesis of CDs do not confer them the desired selectivity, and the pristine material lacks of organic groups for the specific recognition of metals or biological systems. For this reason further modification strategies have had an important role for the developing of novel CDs-based materials and a large number of studies in the last two-three years reported the use of functionalized CDs for sensing and bioimaging. A widespread strategy is based on the CDs covalent functionalization, which exploits the chemical reactivity of the organic groups on the CDs surface in order to form the chemical bond with the moieties intended for the molecular recognition. Moreover this approach can tailor or enhance the photoluminescence features of the nanoparticles. 1.6.2. Functionalization through amide bond Among all the types of covalent functionalization reported in literature, the amide bond formation is without any doubt the most common. In fact the precursors employed for the CDs synthesis frequently carry carboxylic acids, amines or they are treated in oxidizing conditions. Therefore a very simple strategy such as the EDC/NHS (N-(3Dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride/N-hydroxysuccinimide ) activation of carboxylic acids provide a useful path for the amide bond formation (Figure 1.18). 24 Figure 1.17. mechanism of the amide bond formation catalyzed by EDC/NHS. Carboxylic acid reacts with 1-Ethyl-3-(3dimethylaminopropyl)carbodiimide hydrochloride (EDC), forming an unstable intermediate (1). The addition of N-Hydroxysuccinimide (NHS) cause the cleavage of the EDC bond and a less labile NHS ester is formed (2). Finally an amine is added (3) and the amide bond is achieved. Furthermore, some studies reported an enhancement of the fluorescence emission after the formation of the amides, thus revealing an additional advantage of this covalent approach. For example Dong et al. activated by EDC/NHS coupling the carboxylic acids of CDs obtained from hydrothermal treatment of glucose and functionalized them with ethylenediamine.100 The amide bond formation improved the QY from 1.3 to 3.0%. CDs sensitivity towards certain metal ions is often high due to the facility with which those can quench the fluorescence, but selectivity must be improved. For example polyamine molecules such as triethylenetetramine (TETA) are good chelants for copper ions. Therefore the group of Yang used the EDC/NHS protocol for attaching TETA on the carboxylic acids of CDs made from citric acid and urea.101 The as-produced material was used for the detection of Cu2+ and GSH (glutathione). In fact, Cu2+ could bind the CDs surface and quench the PL emission. Further addition of GSH can recover it, due to the stronger ability of GSH to bind the copper ions. In this way GSH concentrations of 0.2-175 µM were detected and the modified CDs were also applied in living yeast cells for detecting Cu2+ and GSH. Similar experiments were performed by Chen et al., who prepared CDs from citric acid and ethylenediamine hydrothermal treatment and coupled them with 1,4,8,11 tetraazacyclotetradecane cyclam through the EDC chemistry.80 Cyclam ring was exploited for hosting copper ions and the CDs fluorescence was completely quenched (Figure 1.19). This material was used in HeLa cells for monitoring the concentration of S2ions, which are able to remove the copper from cyclam and recover the fluorescence. In this way a concentration range of 0-15 µM of S2could be detected. 25 Figure 1.18. fluorescence quenching and recovery of cyclam-functionalized CDs, respectively by Cu2+ and S2ions (vials pictures from ref. 80). An example of CDs-based biosensor is the material prepared by Chai et al., who conjugated CDs carrying –COOH moieties with dopamine through catalyzed amide bond formation.102 This material was used to detect the concentration/activity of tyrosinase, an enzyme that catalyzes the oxidation of the dopamine di-hydroxy phenyl ring to the corresponding quinone.103 The dopapine quinone is able to quench the CDs fluorescence as a result of an intramolecular photo-induced charge transfer and, therefore, the decrease in fluorescence emission is proportional to the tyrosinase activity. Zhong et al. attached instead the glycopeptide antibiotic Vancomycin on CDs made from citric acid and urea, also in this case exploiting the presence of carboxylic groups for the conjugation.104 Vancomycin is able to selectively target the bacteria Staphylococcus aureus, because its terminal peptide sequence D-Ala-D-Ala binds specifically the gram-positive bacteria cell walls. Once vancomycin was covalently linked to the CDs, a quenching was measured because of the bacteria presence, which concentration could be therefore quantitatively determined in the range of 3.18 × 105 – 1.59 × 108 cfu mL-1. Similarly, the group of Wang developed a fluorometric essay for the gram-negative bacteria Salmonella typhimurium.105 The sensor was prepared by coupling to the CDs a specific aptamer, able to recognize the bacteria’s outer membrane proteins. The contact of the functionalized CDs with the bacteria induced a measurable loss of PL emission, corresponding to the bacteria concentration. CDs-based sensors for small biomolecules were also prepared resorting to the amide functionalization. Cyclodextrin has a ring-like molecular structure, where the internal part is hydrophobic while the external hydrophilic, thus promoting the host-guest interaction with hydrophobic molecules such as p-nitrophenol, 26 which is a quencher of the CDs emission. Therefore Sun et al. functionalized the CDs with 6aminoethylamino-β-cyclodextrin via EDC/NHS coupling and filled it with p-nitrophenol.106 The as-prepared material showed excellent sensing properties towards cholesterol, which can replace p-nitrophenol as the cyclodextrin guest molecule, causing a proportional enhancement of the CDs fluorescence (Figure 1.20). Figure 1.19. sensing mechanism based on the host-guest interaction of p-nitrophenol and cholesterol with β-cyclodextrin-functionalized CDs (from ref. 106). An equivalent sensing strategy was exploited by Luo et al., that synthetized CDs from citric acid and cysteine and conjugated them with cyclodextrin. Ferrocenylmethyl trimethylammonium iodide was used as a guest molecule in order to quench completely the CDs emission. This system was then successfully employed for the detection of testosterone, which, being less polar then the ferrocene compound, forms better host-guest interaction with cyclodextrin and could quantitatively recover the fluorescence emission. 1.6.3. Other covalent functionalizations Alternative methods to the amide bond formation for the CDs covalent functionalization involve esterification or sulfonylation. For example, Algarra et al. used mercaptosuccinic acid for functionalizing CDs made from lactose, exploiting therefore the alcohol moieties for the further esterification (Figure 1.21).107 The functionalized CDs could selectively sense Ag+ concentrations of 0-30 µM in water, thanks to the ability of the thiol group to form a complex with silver, which is responsible for the static quenching of the CDs. 27 Figure 1.20. Mercaptosuccinic acid in acid conditions is used for the CDs functionalization. The as-produced nanoparticles are selectively quenched by Ag+ ions (ref. 107). Amino-containing CDs can be modified by sulfonylation, i. e. using sulfonyl chloride compounds for the formation of the sulfonamide. In this way, Wang et al. attached 2,4dinitrobenzene, a specific ligand for selenocysteine, to the nanoparticles, resulting in the complete quenching of their fluorescence.108 Selenocysteine was able to restore the fluorescence, cleaving by nucleophilic substitution the 2,4-dinitrobenzene functionalities, while cysteine, homocysteine, glutathione and others aminoacids didn’t show almost any effect on the CDs emission (Figure 1.22). Therefore the as-prepared sensor showed to be capable of selectively imaging selenol in living cells in a linear range of 0.2-30 µM. Figure 1.21. a) sulfonamide bond formation and cleavage on CDs from ref. 108 and fluorescence dependence. b) selectivity towards selenocysteine (Sec). 28 1.7. CDs in photocatalysis and solar cells 1.7.1. introduction PL phenomena in CDs are the most evident sign of interesting underlying properties, including charge separation and donor-acceptor behavior, photocatalytic activity and photon harvesting ability. The early and rapid discovery of all these properties immediately drew the attention from the energy conversion field. Nowadays researchers are widely employing CDs for the preparation of photocatalysts and solar cells layer components, disclosing step by step their potentiality and revealing the extreme versatility and easiness of processing that are peculiar of this material. In this chapter a description of the CDs role in energy application is given, jointly with few examples. 1.7.2. Photocatalysis Thanks to their special optical behavior, as well as to the high water solubility, excellent environmental compatibility and easiness of production, CDs have been exploited for the preparation of several efficient photocatalysts, in fact CDs acceptor ability stabilizes the photo-induced charge separation occurring on other materials, displacing the electrons and therefore delaying the electron-hole recombination. In the work of S.-Y. Kwak et al. CDs prepared from hydrothermal treatment of ascorbic acid were embedded with mesoporous hematite clusters and the photocatalytic degradation of methylene blue was studied.109 The remarkably higher degradation efficiency of the composite with respect to the hematite alone highlighted the role of CDs, which were able to transfer the electron from hematite to the oxygen specie that starts the methylene blue degradation mechanism. A similar behavior is observed in the CDs/CdS heterojunction prepared by H.-L. Zhang et al., which showed improved photocatalytic ability towards p-nitrobenzene reduction. In here, the CDs were prepared by heating L-cysteine at 280 °C for 5 minutes and mixed with CdS quantum dots via an electrophoretic and sequential chemical bath deposition method, achieving the formation of the heterojunction.110 Upon visible light irradiation, electron-hole separation occurs on CdS and electrons are accepted by CDs, which efficiently catalyze the p-nitrobenzene reduction (Figure 1.23). 29 Figure 1.22. illustration of the CDs/CdS heterojunction and its photocatalytic activity towards p-nitrobenzene (from ref. 110). One last example of the versatility of CDs in photocatalysis, the work of R. Chen et al. reports the evaluation of the photocatalytic activity of the monolayered 2D material BiMoO6 mixed in solution with CDs synthetized from hydrothermal treatment of citric acid and ethylenediamine.111 The efficient degradation under visible light of four pollutants, i.e. ciprofloxacin, bisphenol A, tetracycline hydrochloride, and methylene blue was achieved, promoted by the electron acceptor role of CDs. 1.7.3. Solar cells CDs have been extensively employed in solar cell for covering multiple roles, including as electrode components, mixed in the transport layer or acting as photosensitizer and spectrum converter. H.-S. Choi synthetized CDs by microwave irradiation of citric acid and urea.112 The CDs were used as initiators for the dendritic growth of Au, obtaining in this way the material further employed as counter electrode for the ZnO nanowire/CdS/CdSe quantum dotsensitized solar cell. The as-prepared counter-electrode showed better performance with respect to commonly used Au-sputtered counter electrodes, increasing the power conversion efficiency from 3.6% to 5.4%. The higher performance of the CDs/Au counter electrode was ascribed to its much larger surface area than the Au-sputtered counter electrode, and therefore to an increase in the number of electrocatalytic active sites. The group of E. Palomares instead employed the CDs in the preparation of the hole-transport layer of perovskite solar cells.113 CDs were synthetized by hydrothermal treatment of citric acid and p-phenylenediamine. Next they were included in the solar cell fabrication process as hole-transport material on top of the perovskite layer. In fact, the measured HOMO and LUMO energies of CDs were found to be adequate on one side for ensuring the hole transfer and blocking the electron transfer on the 30 other. The as-prepared device showed a power conversion efficiency of 3%, proving the capability of CDs as hole-transporter material. In the study of M. Q. Zhang et al. CDs were used for increasing the harvesting efficiency of the solar cell, widening its absorption spectrum (Figure 1.24).114 In fact, the absorption of CDs prepared from the hydrothermal treatment of ascorbic acid and silane is found in the UV range, while their fluorescence emission is blue. Blue light matches with the response curve of the active layer, consisting of poly(3-hexylthiophene): [6,6]-phenyl-C61-butyric acid methyl ester (P3HT:PCBM), while UV light does not. Therefore the presence of CDs allows the active layer to partially collect also the UV light, increasing the power conversion efficiency of the device. Figure 1.23. photovoltaic device in which the CDs layer acts as spectral converter (from ref. 114). 1.8. References 1. Cayuela, A.; Soriano, M. L.; Carrillo-Carrión, C.; Valcárcel, M., Semiconductor and carbon-based fluorescent nanodots: the need for consistency. Chemical Communications 2016, 52 (7), 1311-1326. 2. Feng, T.; Zhu, S.; Zeng, Q.; Lu, S.; Tao, S.; Liu, J.; Yang, B., Supramolecular CrossLink-Regulated Emission and Related Applications in Polymer Carbon Dots. ACS Applied Materials & Interfaces 2018, 10 (15), 12262-12277. 3. Shah, S. N. A.; Lin, J.-M., Recent advances in chemiluminescence based on carbonaceous dots. Advances in Colloid and Interface Science 2017, 241, 24-36. 4. Xu, Y.; Liu, J.; Gao, C.; Wang, E., Applications of carbon quantum dots in electrochemiluminescence: A mini review. Electrochemistry Communications 2014, 48, 151154. 5. Xu, X.; Ray, R.; Gu, Y.; Ploehn, H. J.; Gearheart, L.; Raker, K.; Scrivens, W. A., Electrophoretic Analysis and Purification of Fluorescent Single-Walled Carbon Nanotube Fragments. Journal of the American Chemical Society 2004, 126 (40), 12736-12737. 31 6. Zhou, J.; Booker, C.; Li, R.; Zhou, X.; Sham, T.-K.; Sun, X.; Ding, Z., An Electrochemical Avenue to Blue Luminescent Nanocrystals from Multiwalled Carbon Nanotubes (MWCNTs). Journal of the American Chemical Society 2007, 129 (4), 744-745. 7. Liu, H.; Ye, T.; Mao, C., Fluorescent Carbon Nanoparticles Derived from Candle Soot. Angewandte Chemie International Edition 2007, 46 (34), 6473-6475. 8. Hu, S.-L.; Niu, K.-Y.; Sun, J.; Yang, J.; Zhao, N.-Q.; Du, X.-W., One-step synthesis of fluorescent carbon nanoparticles by laser irradiation. Journal of Materials Chemistry 2009, 19 (4), 484-488. 9. Sun, Y.-P.; Zhou, B.; Lin, Y.; Wang, W.; Fernando, K. A. S.; Pathak, P.; Meziani, M. J.; Harruff, B. A.; Wang, X.; Wang, H.; Luo, P. G.; Yang, H.; Kose, M. E.; Chen, B.; Veca, L. M.; Xie, S.-Y., Quantum-Sized Carbon Dots for Bright and Colorful Photoluminescence. Journal of the American Chemical Society 2006, 128 (24), 7756-7757. 10. Zhao, Q.-L.; Zhang, Z.-L.; Huang, B.-H.; Peng, J.; Zhang, M.; Pang, D.-W., Facile preparation of low cytotoxicity fluorescent carbon nanocrystals by electrooxidation of graphite. Chemical Communications 2008, (41), 5116-5118. 11. Peng, H.; Travas-Sejdic, J., Simple Aqueous Solution Route to Luminescent Carbogenic Dots from Carbohydrates. Chemistry of Materials 2009, 21 (23), 5563-5565. 12. Qiao, Z.-A.; Wang, Y.; Gao, Y.; Li, H.; Dai, T.; Liu, Y.; Huo, Q., Commercially activated carbon as the source for producing multicolor photoluminescent carbon dots by chemical oxidation. Chemical Communications 2010, 46 (46), 8812-8814. 13. Gonçalves, H.; Jorge, P. A. S.; Fernandes, J. R. A.; Esteves da Silva, J. C. G., Hg(II) sensing based on functionalized carbon dots obtained by direct laser ablation. Sensors and Actuators B: Chemical 2010, 145 (2), 702-707. 14. Anilkumar, P.; Wang, X.; Cao, L.; Sahu, S.; Liu, J.-H.; Wang, P.; Korch, K.; Tackett Ii, K. N.; Parenzan, A.; Sun, Y.-P., Toward quantitatively fluorescent carbon-based “quantum” dots. Nanoscale 2011, 3 (5), 2023-2027. 15. Liu, Y.; Liu, C.-y.; Zhang, Z.-y., Synthesis and surface photochemistry of graphitized carbon quantum dots. Journal of Colloid and Interface Science 2011, 356 (2), 416-421. 16. Sachdev, A.; Matai, I.; Gopinath, P., Implications of surface passivation on physicochemical and bioimaging properties of carbon dots. RSC Advances 2014, 4 (40), 20915-20921. 17. Bourlinos, A. B.; Stassinopoulos, A.; Anglos, D.; Zboril, R.; Karakassides, M.; Giannelis, E. P., Surface Functionalized Carbogenic Quantum Dots. Small 2008, 4 (4), 455458. 18. Bourlinos, A. B.; Stassinopoulos, A.; Anglos, D.; Zboril, R.; Georgakilas, V.; Giannelis, E. P., Photoluminescent Carbogenic Dots. Chemistry of Materials 2008, 20 (14), 4539-4541. 19. Hsu, P.-C.; Chang, H.-T., Synthesis of high-quality carbon nanodots from hydrophilic compounds: role of functional groups. Chemical Communications 2012, 48 (33), 3984-3986. 20. Hu, S.; Tian, R.; Dong, Y.; Yang, J.; Liu, J.; Chang, Q., Modulation and effects of surface groups on photoluminescence and photocatalytic activity of carbon dots. Nanoscale 2013, 5 (23), 11665-11671. 21. Kozák, O.; Datta, K. K. R.; Greplová, M.; Ranc, V.; Kašlík, J.; Zbořil, R., SurfactantDerived Amphiphilic Carbon Dots with Tunable Photoluminescence. The Journal of Physical Chemistry C 2013, 117 (47), 24991-24996. 22. Yin, J.-Y.; Liu, H.-J.; Jiang, S.; Chen, Y.; Yao, Y., Hyperbranched Polymer Functionalized Carbon Dots with Multistimuli-Responsive Property. ACS Macro Letters 2013, 2 (11), 1033-1037. 38 2. ELUCIDATION OF THE RELATIONSHIP BETWEEN POLYMER STRUCTURE AND BLUE FLUORESCENCE OF CARBON DOTS 39 2.1. Abstract The CDs obtained by the employment of carboxylic acid and amine precursors frequently show bright blue fluorescence emission, independent from the excitation wavelength. The polymer structural rigidity was recognized to play a key role in this phenomenon, nevertheless the knowledge of the CDs structure is still inadequate for its comprehension. In this work three synthetic procedures were carried out to obtain blue fluorescent CDs from carboxylic acids and amines. The full elucidation of the chemical structures and their comparison allowed ascribing unequivocally their similar fluorescent behaviors to the presence of a compact and entangled polyamide network. In here, hydrogen bond-mediated intramolecular interactions were found to significantly enhance the polymer conformational rigidity. Density functional theory calculations of this structure confirmed its rigid and compact nature. Moreover a photo-induced intramolecular charge transfer taking place between amide and carboxylic acid moietes was recognized as the mechanism responsible for the fluorescence emission. Finally, the photoinduced charge-transfer processes could easily explain the performance of CDs in applications as revealed in studies on metal ion sensing. 2.2. Introduction The most common synthesis procedures of fluorescent CDs are generally harsh, involving the hydrothermal treatment or microwave irradiation of organic molecules and/or polymers. These conditions are reflected in the heterogeneous structure of the obtained nanoparticles, which are composed by random polymerization products as well as a certain degree of carbonization. Indeed, at high temperature the molecular precursors condensation, dehydration and other random and unpredictable reactions are expected. However, despite their large variety and uncertain chemical structure, most of the as-produced CDs reveal a unique blue fluorescence emission as common unifying feature. In the last years, some authors related this excitation-independent type of emission to the presence of conjugated molecular fluorophores that are forming on the polymer skeleton during the synthesis process.1-7 Neverthless, B. Yang et al. focused their attention on the non-conjugated polymer structure itself, claiming its key-role in the CDs fluorescence. In fact it is known that also non-conjugated polymers can display fluorescence in condition of high structural rigidity. Here the rigidity of the polymer network, achieved by cross-linking, aggregation or immobilization decreases the vibrational and rotational freedom of subfluorophores such as 40 C=O, N=O, C=N heteroatom-containing double bonds, facilitating their radiative relaxation. This process goes by the name of cross-link enhanced emission (CEE) effect. 8-13 Unfortunately, the difficulty to assign a precise chemical structure to CDs and the major role attributed to the presence of conjugated features diverted the bulk of the attention from the polymer contribution to the CDs fluorescence. The following study proposes a deep investigation on the origin of the excitation-independent blue fluorescence that is commonly observed in a huge variety of CDs. To this end, model types of blue fluorescent CDs with a well-defined structure, consisting of commonly employed carboxylic acid and amine functional groups are synthesized through microwave irradiation or a novel room-temperature synthesis route, exploiting reactions based on carbodiimide chemistry. All the routes produced highly blue emitting fluorescent CDs, whose comparable structural characterization allowed to determine unequivocally their nonconjugated polymer structure and its direct relationship with the optical properties. Density functional theory (DFT) analysis confirmed the experimental data interpretation and provided the mechanism behind the fluorescent behavior, identifying the organic groups involved. 2.3. Experimental Section 2.3.1. Materials Citric acid anhydrous (≥99.5%), tricarballylic acid (99%), ethylendiamine (99+%), N,N'- diisopropylcarbodiimide (99%), nickel (II) nitrate hexahydrate (98.5%), iron (III) chloride (97%), cobalt (II) nitrate hexahydrate (98%), zinc (II) chloride (≥97%), magnesium (II) nitrate hexahydrate (99%), calcium (II) chloride dihydrate (99%), copper (II) nitrate hemi(pentahydrate) (98%), lead (II) nitrate (≥99%), mercurium (II) nitrate hydrate (98%), were used without further purifications. Dialysis tubes with molecular weight cut-off (MWCO) 0.5-1 KDa were bought from Spectrum Labs. 2.3.2. Characterization techniques  The microwave-assisted reaction was performed in a CEM Discover SP reactor employed in open-batch modality.  Atomic force microscopy (AFM) images were acquired in air under ambient conditions using a NT-MDT Aura NTEGRA instrument operating in tapping mode at 110 kHz resonance with Au tips HA_NC ETALON (10nm curvature radius). Samples 41 were prepared on silica substrates by drop casting of diluite water solutions. Particle height distribution analysis was carried out by using the Gwyddion software.  Size exclusion chromatography was performed on a system composed by: pump (Izasa Scientific), automatic injector (Izasa Scientific), PL aquagel column OH-mixedH (Agilent) and refractive index detector T-REX (Wyatt Technology). The mobile phase (0.1 M sodium acetate) was flowed at 1ml/min at 35°C. For the calibration, PEO/PEG polymers were used in the range of 1970-44400 g/mol. The samples were dissolved in the mobile phase at concentration around 1mg/ml. The DLS measurements were recorded on a Malvern Nano Zetasizer HT, on a 10 mm pathlength plastic cuvette.  Elemental analyses were performed in a Thermo Flash EA 1112 instrument with ∼3 mg of powder samples.  Infrared absorption measurements were performed on powder samples pressed with KBr into pellets with a Bruker Vertex 70 spectrometer.  1H and 13C NMR spectra were recorded in D2O solutions at 25 oC on a Bruker AV500 spectrometer (δ in ppm and J in Hz) at a 1H NMR operating frequency of 500.13 MHz. 1H and 13C NMR spectra were referenced using the solvent signal as an internal standard. The assignment of the 1H NMR signals and the corresponding 13C NMR peaks was carried out using standard 1H−13C edited-HSQC and 1H−13C HMBC (JHC = 8 Hz) 2D-experiments. The determination of the diffusion coefficients D (m2/s) was performed at controlled temperature (300 K) in spinning solutions of the corresponding compounds in D2O (concentrations about 2 mM). The values of delta (δ) and delta (Δ) were optimized for each sample. In the case of δ the values found were in the range 1.7-2.0 ms, while for Δ the optimized values ranged from 0.17-0.20 s.  X-ray photoelectron spectroscopy (XPS) measurements were taken with an ESCAPlus spectrometer using a Mg anode (1253.5 eV) and a power of 225 W. XPS data analysis was performed with casaXPS software.  UV/Vis absorption spectra were recorded on a Shimadzu UV-2401 PC spectrophotometer.  Photoluminescence excitation and emission spectra were recorded on a Horiba Jobin Yvon Fluoromax-P, slits of excitation and emission at 1 mm. All the spectra were recorded at room temperature using 10 mm path-length quartz cuvette. 42  Pico-second time-resolved fluorescence spectra were measured by the time-correlatedsingle-photon-counting (TCSPC) method on a Nano-Log spectrofluorometer (Horiba JobinYvon), by using a laser diode as an excitation source (NanoLED, 375 nm) and a UV-Vis detector TBX-PMT series (250-850 nm) by Horiba JobinYvon. Lifetimes were evaluated with the DAS6 Fluorescence-Decay Analysis Software. 2.3.3. Synthesis of CDs1-3 2.3.3.1 Synthesis procedures Synthesis of CDs1. 2.0 g of citric acid monohydrate (CA, 9.5 mmol, 1 eq.) were dissolved in 15 mL of ultrapure water. Upon addition of 0.64 mL of ethylenediamine (EDA, 1 equiv.) the solution was heated up to 140°C through microwave irradiation (stirring, open batch), provoking the evaporation of the water. The temperature was kept constant for one minute, after that the irradiation was stopped and the mixture redissolved in 10 mL of water. The same process was repeated two times more, for a total of three minutes at 140°C. The solid product was diluted with ultrapure water, filtrated through a 0.45 µm PTFE membrane and dialyzed against ultrapure water (MWCO = 0.5-1.0 KDa, 3 days, twice a day), yielding in a brownish powder, with a yield in mass of 35%. Synthesis of CDs2. 2.0 g of tricarballylic acid (TA, 11.4 mmol, 1 equiv.) were dissolved in 15 mL of ultrapure water. Upon addition of 0.76 mL of EDA (1 equiv.) the solution was heated up to 140°C through microwave irradiation (stirring, open batch), provoking the evaporation of the water. The temperature was kept constant for 3 minutes, after that the irradiation was stopped and the mixture cooled down. The solid product was diluted with ultrapure water, filtrated through a 0.45 µm PTFE membrane and dialyzed against ultrapure water (MWCO = 0.5-1.0 KDa, 3 days, twice a day), yielding in a brownish powder with a yield in mass of 27%. Synthesis of CDs3. 1.0 g of anhydrous citric acid (5.2 mmol, 1 equiv.) was dissolved in 5.0 mL of DMF. The solution was cooled in an ice bath and 2.6 mL of diisopropyl carbodiimide (DIC, 3 equiv.) were added. Subsequently, 0.35 mL of EDA (1 eq.) in 5 mL of water were slowly added and the reaction was stirred for 30 minutes at room temperature, during which the mixture turns its color from slightly yellow to red. The filtered reaction was stopped by adding slowly during a time of 3 hours a NaOH solution at pH=10, until no formation of DICurea precipitate was observed. The crude was diluted in ultrapure water, filtered and washed 43 with ethyl acetate. During these operations the coupling agent urea byproduct, in the form of a white precipitate, was removed. The water phase was dialyzed against ultrapure water (MWCO = 0.5-1.0 KDa, 3 days, twice a day). The dry product was obtained by freeze-drying, with a yield in mass of 29 wt. %. It appears as a brownish powder. Scheme 2.1. (Top) reaction of EDA with CA through two synthetic pathways to form CDs1 and CDs3. (Bottom) reaction of EDA with TA to form CDs2. 2.3.3.2 Comments to the synthesis procedures  CA and EDA in a ratio 1:1 were chosen as starting material for the synthesis of CDs1, expecting the amide bond formation via condensation between carboxylic acid and amine groups. The employed reaction conditions are the result of a pre-study aimed to investigate the effect of the reaction time on the structural and optical properties of the CDs (see Appendix B).  CDs2 are obtained through an identical procedure, but using TA in place of CA. The reason is that at high temperature the hydroxyl group in CA could in principle lead to unpredictable products. Thus, the replacement of CA with TA ensures that reactions involving the –OH group are excluded, while the plausible formation of the amide bond is unaffected.  Reaction temperature is an additional crucial parameter to keep under control during the synthesis, since high temperatures enable a wider range of unpredictable reaction pathways. For this reason, CDs3 is achieved by amide bond formation at room temperature, selectively catalyzed by a carbodiimide coupling agent. 44  All the water solutions of the three products display blue photoluminescence when irradiated by UV-light. Figure 2.1. Photographs of CDs1, CDs2 and CDs3 in solid and in water solution (0.5 mg/mL), with and without UV irradiation. 45 2.4. Results and discussion 2.4.1. Size determination of CDs1-3 The three samples were analyzed by AFM (Figure 2.2), DLS (Figure 2.3) and DOSY (Figure 2.4, Table 2.1), finding that they are composed of nanoparticles of around 1.0-1.6 nm. Figure 2.2. AFM images of CDs1 (a), CDs2 (b), CDs3 (c) and their respective height distribution (d) in black, blue and red, respectively. The average height is found at around 1 nm for all the samples. Figure 2.3. DLS size distribution of CDs1 (black), CDs2 (blue), CDs3 (red). All the samples show a diameter of about 1 nm. 46 Figure 2.4. DOSY spectra of CDs1, CDs2, and CDs3. The hydrodynamic radius (rH) was calculated through the Stokes-Einstein equation (Annex A, ), and ranges from 1.2 to 1.6 nm for all the three samples (Table 2.1) Table 2.1. Measured diffusion coefficient (D), calculated hydrodynamic radius (rH) and diameter of CDs1, CDs2, and CDs3. All samples show a diameter between 1.2-1.6 nm. 𝐷 (m2/s) rH (nm) Diameter (nm) CDs1 10-9.4 0.6 1.2 CDs2 10-9.4 0.6 1.2 CDs3 10-9.5 0.8 1.6 47 2.4.2. Optical properties of CDs1-3 The optical properties of CDs1, CDs2 and CDs3 were investigated by UV/Vis spectroscopy (Figure 2.5a), steady state photoluminescence spectroscopy (Figure 2.5b,c) and time-resolved photoluminescence spectroscopy (Figure 2.6a). A peculiar absorption band in the range of 350 - 390 nm appears in the UV/Vis spectra of the three types of CDs, which is directly related to the photoluminescence emission. In fact, the maximum emission intensity, found at 445 nm for CDs1 and CDs3 and at 470 nm for CDs2, is achieved by exciting exactly in the same region (360-390 nm depending on the CDs sample, see Figure 2.5b). Remarkably, emission spectra of CDs1, CDs2 and CDs3 taken at different excitation wavelengths do not reveal a wavelength dependency (Figure 2.5c). Although small shifts of the maximum position are found between the different types of CDs, their absorbance, excitation and emission are highly comparable and can be related to the same radiative process. Figure 2.5. (a) UV/Vis, (b) Excitation and emission spectra of CDs1 (black), CDs2 (blue) and CDs3 (red). (c) Emission in CDs1-3 at different excitation wavelengths. 54 Figure 2.12. (a) 1H NMR. (b) APT 13C NMR, (c) 1H-13C HSQC and (d) 1H-13C HMBC spectra of CDs3. (e) one of the possible chain isomers of the CDs2 repetitive unit, with C and H assignation. NMR interpretation for CDs2: 1H NMR. Hb: 3.8-3.3 ppm (t), EDA 3.25 ppm (s), Hc: 3.23.0 ppm (m), Ha: 3.2-2.3 ppm (s or d). APT 13C NMR shows in-phase signals, belonging to secondary or quaternary carbons. APT also shows an antiphased signal, assigned to C3, at about 36 ppm. C1: 184-171 ppm, C2 and C4: 40-32 ppm, C3: 38-34 ppm. 1H-13C HSQC shows the 1J coupling of Ha, Hb and Hc with C2, C4 and C3 respectively. 1H-13C HMBC shows the 2J of Ha with C3, the 2J coupling of Hc with C1 and C2, the 2J and 3J coupling of Ha with C1 and C2, the 3J coupling of Hc with C1 and the 3J coupling of Hb with C1 and C4. 55 Figure 2.13. (a) 1H NMR. (b) APT 13C NMR, (c) 1H-13C HSQC and (d) 1H-13C HMBC spectra of CDs3. (e) one of the possible chain isomers of the CDs3 repetitive unit, with C and H assignation. NMR interpretation for CDs2: 1H NMR. Hb: 4.0-3.2 ppm (m), EDA 3.35 ppm (s), Ha: 3.22.4 ppm (s or d). APT 13C NMR shows in-phase signals, belonging to secondary or quaternary carbons. C1: 182-171 ppm, C2 and C4: 47-35 ppm, C3: 76-71 ppm. 1H-13C HSQC shows the 1J coupling of Ha and Hb with C2 and C4 respectively. 1H-13C HMBC shows the 2J of Ha with C3, the 2J and 3J coupling of Ha with C1 and C2, the 3J coupling of Hb with C1 and C4. N-acyl DIC urea side-product has formed in little amount during the reaction, incorporated into the CDs5 structure (1H NMR: 1.20, 1.32, 6.72 ppm, in APT: 137, 21 ppm). Some of the proton signals of CDs3 are right-shifted with respect to CDs1, probably because of the different charged form due to the final addition of NaOH. 56 Despite the differences in the synthesis methods, the NMR profiles of CDs1 and CDs3 are remarkably similar, showing that in both cases the same type of polymer structure is obtained. The NMR spectra of CDs2 are also highly comparable, taking into account the differences due to the absence of the hydroxyl group. Merging the information obtained from all the NMR experiments, the chemical structure of CDs1, CDs2 and CDs3 is unambiguously identified as a non-conjugated polymer consisting of the product of condensation of CA and EDA for CDs1 and CDs3, and of TA and EDA for CDs2. The comparison of the NMR spectra also proves that unassigned minor features, specific to each one of the employed synthetic route, are not common to all the samples, and thus cannot form the basis of the fluorescent behavior. Furthermore, NMR experiments provided valuable information regarding the conformation of the polymeric dots. In the 1H NMR spectra, the sharp lineshape of the peaks (singlets and triplets) strongly suggests a compact and static structure, in fast motion with respect to the time of response of the technique. The 1H-13C HSQC spectra show that the methylene carbons couple with a rather condensed set of proton signals. The high density of sharp signals is related to the variety of static chemical environments that surrounds these protons and can be explained with the presence of various chain isomers of the repetitive unit that coexist in the polymer, as well as with the existence of different ionized forms. These data highlight the branched and rigid conformation of the polymeric CDs. 2.4.4. DFT and TDDFT calculations In order to achieve a deeper understanding of the structure and photoluminescence of these nanoparticles, DFT calculations have been performed based on the CDs1 and CDs3 polymeric [C8H12O5N2]n units (see Annex B for details). For the analysis of the structural features that may be responsible of the CEE effect, clusters of different size have been studied, in particular the dimer (n = 2), octamer (n = 8) and decamer (n = 10). The latter is a good approximation of the CDs1 real chain, whose molecular mass was measured to be 2300 Da by size exclusion chromatography (SEC/RI, Annex B), while the polymeric unit weights 216 Da. The optimized structures show that the main characteristic is a very intricate network due to the establishment of both intraand intermolecular hydrogen bonds. In Figure 2.14a two dimeric chains (n = 2) are represented and it is clear the degree of entanglement, both within and between chains. In Figure 2.14b is represented one decameric chain (n = 10) that also shows a highly intricate structure due to the intramolecular hydrogen bonding (HB). This feature is reflected in the size of the nanoparticle. The calculated diameters are 1.564 nm and 57 1.670 nm for the octamer and the decamer, respectively, in good agreement with the particle size measured experimentally. Figure 2.14. Optimized molecular structures of (a) two dimer (n = 2) chains and (b) one decamer (n =10) chain. (c) HOMO and (d) LUMO molecular orbitals involved in the fluorescence phenomenon. As it has been previously mentioned, this aggregation of the chains is responsible of the enhancement of the fluorescence in these systems, due to a restriction of the vibrational or rotational degrees of freedom that may favor a non-radiative relaxation. In order to corroborate this statement, the optical properties of four representative systems have been calculated, namely, three dimers and one octamer. The first dimer has only one chain, while the others comprise two dimeric chains to generate two conformers, where either the intramolecular or the intermolecular HB is favored. The use of one or two chains, as well as two different conformers, allows studying the influence of the intraand intermolecular interactions independently. The octamer has been used in order to relieve the computational effort. In Table 2.3 are gathered the absorption and emission energies, and the corresponding wavelengths. 58 Table 2.3. Absorption (Eabs) and emission (Eem) energies Eabs (eV) Eem (eV) Dimer (one chain) 5.1811 --- Dimer of two chains (HB intramolecular) 4.8794 --- Dimer of two chains (HB intermolecular) 5.1382 1.8317 Octamer 4.6142 1.9602 It is observed that the four models absorb in the UV, with wavelengths in the range of abs = 239 – 269 nm. The nature of the transition has been characterized by the analysis of the molecular orbitals involved, which is depicted in Figures 2.14c and 2.14d for the dimer, in order to facilitate the visualization. The HOMO has its main contribution from the amide moiety (-CONH-), while the LUMO is centered in the carboxylic groups (-COOH) and it is from HOMO to LUMO where excitation takes place. This means that HOMO and LUMO molecular orbitals are confined at very specific sites, that they are well separated, and that the fluorescence is a charge transfer process. This process is enhanced for the large models due to the sum of all local charge transfer processes occurring in each amide-carboxyl pair of the chain. Experimental evidence for these findings is provided by the results of our photoluminescence studies. First, the large Stokes-shift of about 1 eV (see Figure 2.5b) should be related to a considerable change in the dipole moment between ground state and excited state, being consistent with the typical photo-induced charge transfer mechanism.54-57 Second, the protonation dependency of the emission intensity (see Figure 2.7) reveals the important role of the carboxylic acids in the fluorescence process: They affect the rigidity of the overall polymer network structure via hydrogen bonding or electrostatic interactions, as discussed above. Concomitantly, they act as electron acceptor in the charge transfer process. The observation of a strong decay of the fluorescence in an acidic environment is a consequence of the weakening of the carboxyl acids intramolecular interactions, which create the emitting state (see Figure 2.14c,d). On the contrary, strengthening the carboxylic intramolecular chain interactions, as obtained under neutral and basic conditions, leads to the recovery of the carboxylic acids intramolecular chain interactions, favoring charge-transfer and the fluorescence process. Regarding the emission, a striking behavior is found. Both the dimer with one chain and that with two chains where the main interaction is the intramolecular HB, show transitions that are 59 remarkably redshifted. In fact, the emission takes place in a region where the singlet excited state and the ground state potential energy surfaces cross, known as internal conversion (IC), involving a non-radiative relaxation and, hence, no fluorescence will be observed. Considering the dimer with two chains, in which the intermolecular HB is favored, a clear redshift of the emission energy is observed with respect to the absorption, arising from a singlet excited state well separated from the ground state (1.8317 eV above the ground state). This emission takes place in the visible region (em = 676.9 nm). These results suggest that, in a small chain like a dimer, the crosslink generated by the intramolecular HB is not enough to avoid the non-radiative relaxation by vibration or rotation of the chain, and the main confinement is exerted by the intermolecular interactions. Nevertheless, this prominent feature disappears when longer chains are considered and a larger number of intramolecular interactions are allowed. Thus, inspecting the results for the octamer (single chain), a similar redshift is observed and the emission also take place in the visible (em = 632.5 nm). This means that the intramolecular crosslink is strong enough to hinder the vibration and rotation of the chain, and fluorescence is observed. At this point it is worthwhile to outline that the theoretical calculations are based on simplified models, which do not take into account several effects of experimental relevance, such as polymer branching, influence of additional chains, solvent effects, nor those related to the presence of free amines or charged states. These parameters contribute to enhance the rigidity of the polymer and thus may lead to further important down shifts of the calculated emission wavelengths towards the blue wavelength range. Importantly, despite quantitative shortfalls, our models yet provide a solid base to qualitatively explain the origin of the fluorescence of the CDs. 2.4.5. Interpretation of the CDs sensing ability The identification of charge-transfer processes between well-defined and spatially separated functional groups as origin of the fluorescence in polymer CDs now provides a powerful toolbox towards an improved understanding of their use in technological applications. This is demonstrated at hand of studies on the sensing of metal ions in aqueous solutions. To this end, the fluorescence of an aqueous solution of CDs3 (0.4 mg/mL) was measured in the presence of several types of metal ions at a concentration of 1mM. A high affinity (i.e. high quenching rate) towards Fe3+, Cu2+, and Hg2+, and other types of metal ions to a lower content is observed (Figure 2.15). Similar results were obtained for CDs1 and CDs2, which are also in agreement with those observed for CDs of unknown chemical structure.58-60 These findings 60 clearly confirm that the high sensitivity can be ascribed to the common chemical structure of polymer CDs. In fact, the same groups identified to be involved in the charge transfer process can act also as chelating agents, in a similar way to EDA, EDTA and citrate. Carboxylic acids, amides and amines are therefore interacting strongly with the metal ions and the binding provides favorable non-radiative relaxation pathways, which compete with the intramolecular charge-transfer process and drastically reduce the photoluminescence emission. This study not only proves the existence of photoinduced charge-transfer processes between spatially separated amide and carboxylic moieties in polymer CDs, but also explains how to successfully exploit this mechanism in other types of applications of technological relevance. Figure 2.15. (a) Relative fluorescence intensity of CDs3 where I and Io are the intensities in presence and absence of the metal ions. (b) Emission spectra of a CDs3 water solution in the presence of 1mM concentration of Mg2+, Ca2+, Fe3+, Pb2+, Ni2+, Cu2+, Ag2+, Zn2+, Co2+, Hg2+. 61 2.5. Conclusion The synthesis of blue fluorescent model polymer CDs was obtained from CA and EDA as well as from TA and EDA by microwave irradiation, and from CA and EDA by a novel coupling agent-mediated condensation at room temperature. The optical features of the three types of model CDs of 1 – 1.5 nm in size are remarkably similar, suggesting that the copolymerization of the reagents by amide bond is a sufficient condition for the formation of the fluorophore in polymer CDs. This hypothesis was further confirmed by a detailed structural characterization of the materials, which revealed that the presence of H-bonds and electrostatic interactions goes along with the high conformational rigidity of the polymeric chain. DFT and TDDFT calculations of the octamer of this structure indeed show that supramolecular H-bond mediated interactions cause the rigid entanglement of the chains, hindering vibration and rotation, and facilitate the radiative relaxation process. Finally, the amide and the carboxylic groups are recognized as the moieties, which respectively provide the main contribution to the HOMO and the LUMO molecular orbitals. Photoinduced chargetransfer between these spatially separated groups, assisted by H-bond mediated supramolecular interactions in the entangled polyamide network therefore is identified as origin of the fluorescence phenomena in polymer carbon dots. Figure 2.16. Illustration of the CDs formation and of the photo-induced charge transfer phenomenon. It is proposed that this single emitter type process provides a universal explanation for the CDs’ blue fluorescence observed in the vast and increasing amount of reports where these 62 nanoparticles are prepared from carboxylic acids and amines. However, this process becomes masked by the contribution of additional fluorescent emitters, typically obtained when employing harsher conditions, resulting in an excitation dependent emission behavior. Additionally these findings explain the performance of polymer CDs in existing technological applications. Moreover, it paves the way to exploit novel synthetic bottom-up routes to obtain polymer CDs with tailored properties uncovering yet unexploited opportunities. 2.6. References 1. Song, Y.; Zhu, S.; Zhang, S.; Fu, Y.; Wang, L.; Zhao, X.; Yang, B., Investigation from chemical structure to photoluminescent mechanism: a type of carbon dots from the pyrolysis of citric acid and an amine. Journal of Materials Chemistry C 2015, 3 (23), 5976-5984. 2. Zhu, S.; Zhao, X.; Song, Y.; Lu, S.; Yang, B., Beyond bottom-up carbon nanodots: Citric-acid derived organic molecules. Nano Today 2016, 11 (2), 128-132. 3. Dhenadhayalan, N.; Lin, K.-C.; Suresh, R.; Ramamurthy, P., Unravelling the Multiple Emissive States in Citric-Acid-Derived Carbon Dots. The Journal of Physical Chemistry C 2016, 120 (2), 1252-1261. 4. Schneider, J.; Reckmeier, C. J.; Xiong, Y.; von Seckendorff, M.; Susha, A. S.; Kasák, P.; Rogach, A. L., Molecular Fluorescence in Citric Acid-Based Carbon Dots. The Journal of Physical Chemistry C 2017, 121 (3), 2014-2022. 5. Shi, L.; Yang, J. H.; Zeng, H. B.; Chen, Y. M.; Yang, S. C.; Wu, C.; Zeng, H.; Yoshihito, O.; Zhang, Q., Carbon dots with high fluorescence quantum yield: the fluorescence originates from organic fluorophores. Nanoscale 2016, 8 (30), 14374-14378. 6. Fang, Q.; Dong, Y.; Chen, Y.; Lu, C.-H.; Chi, Y.; Yang, H.-H.; Yu, T., Luminescence origin of carbon based dots obtained from citric acid and amino group-containing molecules. Carbon 2017, 118, 319-326. 7. Ehrat, F.; Bhattacharyya, S.; Schneider, J.; Löf, A.; Wyrwich, R.; Rogach, A. L.; Stolarczyk, J. K.; Urban, A. S.; Feldmann, J., Tracking the Source of Carbon Dot Photoluminescence: Aromatic Domains versus Molecular Fluorophores. Nano Letters 2017, 17 (12), 7710-7716. 8. Zhu, S.; Song, Y.; Zhao, X.; Shao, J.; Zhang, J.; Yang, B., The photoluminescence mechanism in carbon dots (graphene quantum dots, carbon nanodots, and polymer dots): current state and future perspective. Nano Research 2015, 8 (2), 355-381. 9. Zhu, S.; Song, Y.; Shao, J.; Zhao, X.; Yang, B., Non-Conjugated Polymer Dots with Crosslink-Enhanced Emission in the Absence of Fluorophore Units. Angewandte Chemie International Edition 2015, 54 (49), 14626-14637. 10. Zhao, E.; Lam, J. W. Y.; Meng, L.; Hong, Y.; Deng, H.; Bai, G.; Huang, X.; Hao, J.; Tang, B. Z., Poly[(maleic anhydride)-alt-(vinyl acetate)]: A Pure Oxygenic Nonconjugated Macromolecule with Strong Light Emission and Solvatochromic Effect. Macromolecules 2015, 48 (1), 64-71. 11. Qiao, Z. A.; Huo, Q.; Chi, M.; Veith, G. M.; Binder, A. J.; Dai, S., A "ship-in-abottle" approach to synthesis of polymer dots@silica or polymer dots@carbon core-shell nanospheres. Adv Mater 2012, 24 (45), 6017-21. 12. Tao, S.; Song, Y.; Zhu, S.; Shao, J.; Yang, B., A new type of polymer carbon dots with high quantum yield: From synthesis to investigation on fluorescence mechanism. Polymer. 63 13. Tao, S.; Lu, S.; Geng, Y.; Zhu, S.; Redfern, S. A. T.; Song, Y.; Feng, T.; Xu, W.; Yang, B., Design of Metal-Free Polymer Carbon Dots: A New Class of Room-Temperature Phosphorescent Materials. Angewandte Chemie International Edition 2018, 57 (9), 23932398. 14. Xiao, Q.; Liang, Y.; Zhu, F.; Lu, S.; Huang, S., Microwave-assisted one-pot synthesis of highly luminescent N-doped carbon dots for cellular imaging and multi-ion probing. Microchimica Acta 2017, 184 (7), 2429-2438. 15. Gao, F.; Ma, S.; Li, J.; Dai, K.; Xiao, X.; Zhao, D.; Gong, W., Rational design of high quality citric acid-derived carbon dots by selecting efficient chemical structure motifs. Carbon 2017, 112, 131-141. 16. Zhang, Y.; Wang, Y.; Feng, X.; Zhang, F.; Yang, Y.; Liu, X., Effect of reaction temperature on structure and fluorescence properties of nitrogen-doped carbon dots. Applied Surface Science 2016, 387, 1236-1246. 17. Xu, Z.-Q.; Lan, J.-Y.; Jin, J.-C.; Dong, P.; Jiang, F.-L.; Liu, Y., Highly Photoluminescent Nitrogen-Doped Carbon Nanodots and Their Protective Effects against Oxidative Stress on Cells. ACS Applied Materials & Interfaces 2015, 7 (51), 28346-28352. 18. He, J.; Zhang, H.; Zou, J.; Liu, Y.; Zhuang, J.; Xiao, Y.; Lei, B., Carbon dots-based fluorescent probe for “off-on” sensing of Hg(II) and I−. Biosensors and Bioelectronics 2016, 79, 531-535. 19. Ogi, T.; Aishima, K.; Permatasari, F. A.; Iskandar, F.; Tanabe, E.; Okuyama, K., Kinetics of nitrogen-doped carbon dot formation via hydrothermal synthesis. New Journal of Chemistry 2016, 40 (6), 5555-5561. 20. Zhang, F.; Feng, X.; Zhang, Y.; Yan, L.; Yang, Y.; Liu, X., Photoluminescent carbon quantum dots as a directly film-forming phosphor towards white LEDs. Nanoscale 2016, 8 (16), 8618-8632. 21. Rong, M.; Feng, Y.; Wang, Y.; Chen, X., One-pot solid phase pyrolysis synthesis of nitrogen-doped carbon dots for Fe3+ sensing and bioimaging. Sensors and Actuators B: Chemical 2017, 245, 868-874. 22. Bhattacharyya, S.; Ehrat, F.; Urban, P.; Teves, R.; Wyrwich, R.; Döblinger, M.; Feldmann, J.; Urban, A. S.; Stolarczyk, J. K., Effect of nitrogen atom positioning on the tradeoff between emissive and photocatalytic properties of carbon dots. Nature Communications 2017, 8 (1), 1401. 23. Bagheri, Z.; Ehtesabi, H.; Rahmandoust, M.; Ahadian, M. M.; Hallaji, Z.; Eskandari, F.; Jokar, E., New Insight into the Concept of Carbonization Degree in Synthesis of Carbon Dots to Achieve Facile Smartphone Based Sensing Platform. Scientific Reports 2017, 7 (1), 11013. 24. Tong, G.; Wang, J.; Wang, R.; Guo, X.; He, L.; Qiu, F.; Wang, G.; Zhu, B.; Zhu, X.; Liu, T., Amorphous carbon dots with high two-photon fluorescence for cellular imaging passivated by hyperbranched poly(amino amine). Journal of Materials Chemistry B 2015, 3 (4), 700-706. 25. Liu, Y.; Zhou, L.; Li, Y.; Deng, R.; Zhang, H., Facile synthesis of nitrogen-doped carbon dots with robust fluorescence in a strongly alkaline solution and a reversible fluorescence 'off-on' switch between strongly acidic and alkaline solutions. RSC Advances 2016, 6 (110), 108203-108208. 26. Yang, M.; Li, B.; Zhong, K.; Lu, Y., Photoluminescence properties of N-doped carbon dots prepared in different solvents and applications in pH sensing. Journal of Materials Science 2018, 53 (4), 2424-2433. 27. Zhang, W.; Shi, L.; Liu, Y.; Meng, X.; Xu, H.; Xu, Y.; Liu, B.; Fang, X.; Li, H.-B.; Ding, T., Supramolecular interactions via hydrogen bonding contributing to citric-acid 70 with the methylenes in the 1H-13C HMBC spectra but not in the 1H-13C HSQC spectra. Gathering all these structural information, it appears clear that all the four methods employed accomplished the synthesis of polymer nanoparticles whose structure consists of the condensation products between CA carboxylic acids and EDA amines. In spite of the structural similarity, an investigation of the CDs 1a-1d optical properties showed that the fluorescence intensity depends considerably on the synthetic method. UV/Vis spectra of CDs 1a, CDs 1b and CDs 1d present an absorption band at 340 nm, while for CDs 1c the same feature has its maximum at 310 nm (Figure 3.1a). Probably in CDs 1c the shift is not real, due to the fact that the considered absorption band is very small with respect to the absorption of the C=O groups at <300 nm. The emitting behavior of the four samples shows indeed a correlation with the quantum yield (QY). In CDs 1a (QY=17.7%) and 1b (QY=8.6%) the emission takes place around 445 nm and is approximately independent from the excitation, while in CDs 1c and 1d the QY is remarkably lower (respectively 2.2% and 4.0%) and the emission is excitation-dependent (Figure 3.1b). These observations suggest that the main contribution to the fluorescence intensity is independent from the excitation. Additionally some effects produce the excitation dependency in all the samples. However, their influence only becomes noticeable when the main absorption contribution is weak, as in the case of CDs 1c and 1d. The excitation dependency in these cases could be ascribed to heterogeneity of the emitting centers, possibly due to crosslinking or aggregation effects14. 71 Figure 3.1. a) absorption spectra of CDs 1a-d. b) emission of CDs 1a-d for different excitation wavelengths. Once attested that the DICand EDC-mediated condensation are, among the methods considered, the most effective for the synthesis of highly fluorescent nanoparticles from CA and EDA, the same procedure was tested employing different aliphatic and aromatic polyamines in the place of EDA, again in a ratio 1:1 with respect to CA. In this way the synthesis of CDs 2a-f was accomplished (Scheme 3.2, Table 3.1). 72 Scheme 3.2. Synthesis of CDs 2a-f. Table 3.1. amines employed for the polycondensation, emission maximum, QY and pictures (in water, UV light off and on) of CDs 2a-f. The as-produced materials CDs 2a-f were characterized by AFM (Figure 9.32) and DOSY (Figure 9.33 and Table 9.3). Accordingly to both techniques, all the samples consist of nanoparticles of size comprised between 1.0 and 2.5 nm, with the exception of CDs 2c, which shows a size of around 2 nm by AFM, but of 4.8 nm by DOSY. The as-produced materials were characterized by elemental analysis, IR and NMR spectroscopy. Also for CDs 2a-c, the C, H, N and O percentages strongly suggest that the formed copolymers are made of a ratio 1:1 between CA and the employed amine (Tables 9.4-9.6 and Figures 9.34-9.36), while this stoichiometry is not respected for CDs 2d-f, presumably due to the lower and co-dependent reactivity of the aromatic di-amines (Table 9.7 and Figure 9.37), but also to the presence of unknown impurities (Figures 9.56, 9.60, 9.64). IR spectra (Figures 9.38-9.43) show C=O 73 stretching for carboxylic acids and amides, as previously observed for CDs 1a-d. Moreover CDs 2d-f present the typical aromatic C=C stretching between 1516-1497 cm-1 (Figures 9.419.43). Finally, 1H, 13C APT, 1H-13C HSQC and 1H-13C HMBC experiments confirmed the copolymer connectivity via amide as the results of the condensation between citric acid and the respective amine (Figures 9.44-9.67). For CDs 2d-f are also well visible the aromatic signals, between 7.5-6.5 ppm in the proton spectra and 147-109 ppm in the carbon spectra. The fact that the results of the structural characterization conclusively confirm the expected polyamide structures emphasizes the versatility of our synthetic procedure. Concerning the optical properties of CDs 2a-f, it can be observed that the emission wavelength does not change consistently with the amine employed and it is found between 433 and 452 nm depending on the sample (Figures 9.69-9.74). On the other hand the QY of CDs 2a-c, with aliphatic amines, is higher than in CDs 2d-f, with aromatic amines (Table 3.1). It is reasonable to say that the polymer conformation responsible for the radiative decay can be hampered by the locked configuration of the amines in o-, mor pposition. Among these, oposition could be the most favorable, resembling one of the EDA conformations, and in fact CDs 2d QY is higher than CDs 2e QY, which in turn is higher than the one of CDs 2f, whose substituents do not allow the compact folding needed for the formation of the fluorophore. The finding that the polycondensation catalyzed by carbodiimides can be successfully exploited for the synthesis of different fluorescent nanoparticles as a function of the type of amine group employed, allowing thus the control on the final chemical structure, provided a unique opportunity to perform an additional study testing the possibility to functionalize insitu the as-produced materials. Since the growth of the nanoparticles is mediated by the activation of the carboxylic acids via coupling agents, it can be interrupted at any moment adding an excess of a strong nucleophile. In the previous experiments, NaOH was used to restore the carboxylic acids and quench the polymerization. Nevertheless, primary amines are equally good candidates for this purpose. Moreover, the employment of an amine offers the possibility to attach onto the surface of the CDs virtually any moiety that is stable and soluble in the reaction medium. The in-situ functionalization was tested by the addition of five different primary amines during the polycondensation reaction between CA and EDA mediated by DIC (Scheme 3.3, Table 3.2). 74 Scheme 3.3. a) polycondensation of CA and EDA mediated by coupling agent. b) the addition of a primary amine consumes the activated carboxylic acids and stops the polymerization. Table 3.2. amine employed for the functionalization, emission maximum, QY and pictures (in water, UV light off and on) of CDs 3a-e. Similarly to CDs 1a-d, CDs 3a-e reveal sizes between 1.0-2.6 nm, as probed by AFM (Figure 9.76) and DOSY (Figure 9.77, Table 9.8). The structural characterization proved the effective functionalization. Elemental analysis shows that the functionalized materials CDs 3a-e present lower molar percentages of oxygen compared to CDs 1a, as a consequence of the successful functionalization (Table 9.9 and Figure 9.78). In parallel, in the infrared spectra of CDs 3a-3d (Figures 9.79-9.83) the carboxylic acid C=O stretching at 1704 cm-1 is weaker than in CDs 1a or not visible, due to the consumption of the majority of the carboxyl groups by means of the functionalization. In CDs 3e this effect is not appreciable because of the 75 presence of the carbamate C=O stretching band, which is found at similar wavenumbers of the carboxylic one. Additionally it is observed in CDs 3c the C=C stretching at 1545 cm-1 due to the aromatic ring. 1H, 13C APT, 1H-13C HSQC and 1H-13C HMBC experiments confirmed the introduction on CDs 3a-3e of the corresponding functionalities (Figures 9.84-9.103). The performed structural modifications only slightly affect the absorption (Figure 9.104) and emission maximum position (Figures 9.105-9.109), meaning that they are not directly involved in the fluorescence mechanism. However different QY are obtained depending on the functionalization. Rigid and bulky groups, such as in CDs 2c and 3d, decrease the fluorescence intensity. In contrast, the QY benefits from the thiol functionalization, possibly due to its involvement in the H-bond interactions that play a role in the formation of the polymer conformation needed for the obtaining of the fluorescent behavior. 3.5. Conclusion In summary, we illustrated the versatility of a novel approach for the synthesis of fluorescent CDs, consisting in the coupling agent-mediated condensation between citric acid and amines at room temperature. This strategy enables the synthesis of a wide variety of polymeric nanoparticles whose chemical structure is determined only by the choice of the reagents. Furthermore, the method allows the in-situ functionalization of the CDs, supplying an additional level of control on the final chemical structure and avoiding further functionalization steps otherwise needed for many applications. Therefore this method provides an easy and efficient pathway for controlling the structure and the chemical properties of the polymer nanoparticles, furnishing an extremely versatile tool for the design of CDs-based material such as sensors and drug nanocarriers. 3.6. References 1. Yan, F.; Jiang, Y.; Sun, X.; Bai, Z.; Zhang, Y.; Zhou, X., Surface modification and chemical functionalization of carbon dots: a review. Microchimica Acta 2018, 185 (9), 424. 2. Fu, C.; Qian, K.; Fu, A., Arginine-modified carbon dots probe for live cell imaging and sensing by increasing cellular uptake efficiency. Materials Science and Engineering: C 2017, 76, 350-355. 3. Chai, L.; Zhou, J.; Feng, H.; Tang, C.; Huang, Y.; Qian, Z., Functionalized Carbon Quantum Dots with Dopamine for Tyrosinase Activity Monitoring and Inhibitor Screening: In Vitro and Intracellular Investigation. ACS Applied Materials & Interfaces 2015, 7 (42), 23564-23574. 76 4. Zhong, D.; Zhuo, Y.; Feng, Y.; Yang, X., Employing carbon dots modified with vancomycin for assaying Gram-positive bacteria like Staphylococcus aureus. Biosensors and Bioelectronics 2015, 74, 546-553. 5. Krishna, A. S.; Radhakumary, C.; Antony, M.; Sreenivasan, K., Functionalized carbon dots enable simultaneous bone crack detection and drug deposition. Journal of Materials Chemistry B 2014, 2 (48), 8626-8632. 6. Zhang, J.; Zhao, X.; Xian, M.; Dong, C.; Shuang, S., Folic acid-conjugated green luminescent carbon dots as a nanoprobe for identifying folate receptor-positive cancer cells. Talanta 2018, 183, 39-47. 7. Yang, L.; Wang, Z.; Wang, J.; Jiang, W.; Jiang, X.; Bai, Z.; He, Y.; Jiang, J.; Wang, D.; Yang, L., Doxorubicin conjugated functionalizable carbon dots for nucleus targeted delivery and enhanced therapeutic efficacy. Nanoscale 2016, 8 (12), 68016809. 8. Yang, R.; Guo, X.; Jia, L.; Zhang, Y., A fluorescent “on-off-on” assay for selective recognition of Cu(II) and glutathione based on modified carbon nanodots, and its application to cellular imaging. Microchimica Acta 2017, 184 (4), 1143-1150. 9. Chen, J.; Li, Y.; Lv, K.; Zhong, W.; Wang, H.; Wu, Z.; Yi, P.; Jiang, J., Cyclamfunctionalized carbon dots sensor for sensitive and selective detection of copper(II) ion and sulfide anion in aqueous media and its imaging in live cells. Sensors and Actuators B: Chemical 2016, 224, 298-306. 10. Luo, M.; Hua, Y.; Liang, Y.; Han, J.; Liu, D.; Zhao, W.; Wang, P., Synthesis of novel β-cyclodextrin functionalized S, N codoped carbon dots for selective detection of testosterone. Biosensors and Bioelectronics 2017, 98, 195-201. 11. Mohammadi, S.; Salimi, A.; Hamd-Ghadareh, S.; Fathi, F.; Soleimani, F., A FRET immunosensor for sensitive detection of CA 15-3 tumor marker in human serum sample and breast cancer cells using antibody functionalized luminescent carbon-dots and AuNPs-dendrimer aptamer as donor-acceptor pair. Analytical Biochemistry 2018, 557, 18-26. 12. Sun, Q.; Fang, S.; Fang, Y.; Qian, Z.; Feng, H., Fluorometric detection of cholesterol based on β-cyclodextrin functionalized carbon quantum dots via competitive host-guest recognition. Talanta 2017, 167, 513-519. 13. Vallan, L.; Urriolabeitia, E. P.; Ruipérez, F.; Matxain, J. M.; Canton-Vitoria, R.; Tagmatarchis, N.; Benito, A. M.; Maser, W. K., Supramolecular-Enhanced Charge Transfer within Entangled Polyamide Chains as the Origin of the Universal Blue Fluorescence of Polymer Carbon Dots. Journal of the American Chemical Society 2018. 14. Feng, T.; Zhu, S.; Zeng, Q.; Lu, S.; Tao, S.; Liu, J.; Yang, B., Supramolecular Cross-Link-Regulated Emission and Related Applications in Polymer Carbon Dots. ACS Applied Materials & Interfaces 2018, 10 (15), 12262-12277. 77 4. ELECTRONIC INTERACTIONS IN CDs/MoS2 ELECTROSTATIC COMPLEX 78 4.1. Abstract In this chapter, a study on the electronic interactions between carbon dots (CDs) and functionalized molybdenum disulfide (MoS2) oligolayers is reported. For this purpose, a solution of negatively charged CDs obtained from the microwave irradiation of citric acid and ethylenediamine was titrated with ammonium-functionalized positively charged MoS2, and the optical properties of the forming electrostatic complex were studied ongoing. Efficient fluorescence quenching of CDs by MoS2 was observed and attributed to photoinduced electron/energy transfer as the decay mechanism for the transduction of the singlet excited state of CDs. Finally, the CD/MoS2 complex was tested as catalyst towards the hydrogen evolution reaction and found to be superior to that of individual CDs species. 4.2. Introduction MoS2 is a member of the layered-transition metal dichalcogenides (TMDs), a group to which belong also tungsten disulfide (WS2), tantalum(IV) sulfide (TaS2), titanium disulfide (TiS2), tungsten diselenide (WSe2), molybdenum diselenide (MoSe2), and many others.1 These materials consist of pseudo-2D crystal layers held together by non-covalent interactions. The number of layers is directly related to the TMDs optical and electronic properties, which therefore can be tailored as a result of an exfoliation process of the bulk material. For example, bulk MoS2 shows semi-conducting properties, having a band gap of 1.3 eV. Depending on the exfoliation process, oligolayered or monolayered MoS2 can be obtained, and decreasing the number of layers the band gap increases, up to 1.8 eV for the single layered material. Furthermore, the width of the band gap is directly related to the optical properties of the TMDs: increasing the bandgap energy in MoS2 results in changes in photoconductivity, absorption and photoluminescence. From bulk to monolayer, MoS2 shows an increase of photoluminescence quantum yield by a factor of up to 104.2 Clearly, the photoelectronic response of MoS2 can be of great interest for the design of optoelectronic devices, such as photo-transistor3, diodes4, chemical sensors5 and LEDs.6 Moreover, the direct bandgap energy is located in the visible range, allowing the use of MoS2 in solar cells and energy conversion.7-9 However, just like many 2D materials, the processing of MoS2 has some draw-backs. In fact, the MoS2 single layers are hardly dispersible and once in solution they are not well stabilized by the solvent, thus they rapidly aggregate 79 and precipitite as a result of the energetically favorable stacking between layers. To overcome this problem, several approaches, both covalent and non-covalent functionalization with organic molecules, were explored. For example, the group of Tagmatarchis achieved the covalent bonding of 1,2-dithiolanes on the edges of semiconducting MoS2 under mild conditions.10 In here, dithiolanes are reacting with the MoS2 edges, where the reactive sulphur vacancies are mostly found, thus preserving the basal plane optoelectronic properties. Furthermore, this method allows introducing charged organic groups, such as ammonium salts, on the MoS2 layers, improving on one side the dispersibility in polar solvents and on the other side providing suitable moieties for the facile coupling with other photoactive materials via electrostatic interactions. In this way, hybrid systems with new optoelectronic properties can be easily prepared. In this work, the photophysical and electrocatalytic properties of CDs/MoS2 complexes are investigated. CDs bearing carboxylic acids were prepared by a standard microwave irradiation protocol from citric acid and ethylenediamine. Afterwards, the negatively charged carboxylate form was obtained by mild alkaline treatment. In parallel, positively charged MoS2 monolayers were prepared in a two-steps procedure, involving first the covalent functionalization of exfoliated MoS2 with 1,2-dithiolane tert-butyl carbamate and second its acidic deprotection yielding the positively charged ammonium salt. Next, the titration of the negative CDs with the positive MoS2 was followed by UV/vis spectroscopy, steady-state and time-resolved photoluminescence spectroscopy, allowing to scrutinize the intraensemble electronic interactions between the two species. Finally, overpotentials and Tafel slopes were evaluated for the assessment of the electrocatalytic activity of CD/MoS2 toward the hydrogen evolution reaction (HER). 4.3. Experimental section 4.3.1. Materials Molybdenum disulfide powder (>99%), chlorosulfonic acid (99%), citric acid anhydrous (≥99.5%) and ethylendiamine (99+%) were used without further purifications. Dialysis tubes with molecular weight cut-off (MWCO) 0.5-1 KDa were bought from Spectrum Labs. 86 4.3.3.3 Characterization of MoS2-based materials 2 and 3 Figure 4.6. ATR-IR spectra for 1,2-dithiolane derivative 1 (red) and MoS2-based materials 2 (black) and 3 (blue). ATR-IR spectra comparison of materials 2 and 3 furnish a proof for the deprotection step (figure 4.6). While stretching vibration bands due to C-H units are identified in the region 2800-3000 cm-1 for both 2 and 3, two discrete bands at 1650 and 1710 cm-1 owed to carbonyl amide and BOC units, respectively, are present in the IR spectrum of 2, with the latter band being absent in the spectrum of 3, thus justifying the effective removal of BOC. 87 Figure 4.7. Normalized Raman spectra for exfoliated MoS2 (black) and MoS2-based materials 2 (gray) and 3 (blue), obtained upon 514 nm excitation. Raman spectroscopy revealed the presence of the characteristic A1g and E12g modes located at 406 and 382 cm-1, respectively, in materials 2 and 3 (Figure 4.7). Moreover, the A1g and E12g modes were found unaltered as compared with the ones present in exfoliated MoS2. Since for exfoliated MoS2 the calculated frequency difference between A1g and E12g is 24 cm-1, corresponding to the presence of 3-4 MoS2 layers in average11, it is reasonable to claim that the same number of layers exists in 2 and 3. Additionally, no other Raman bands were observed in the region 500-1000 cm-1, indicating the absence of oxidation during the exfoliation and functionalization process, hence, proving the preservation of the electronic properties of the semiconducting MoS2 polytype. 88 Figure 4.8. Thermographs for MoS2-based material 3. The ζ-potential value changed from -24 mV for exfoliated MoS2 to +2.6 mV for 3, being consistent with the presence of ammonium functionalities. Moreover, Kaiser test revealed a value of 50 μmol/g for free amine units in 3. Then, based on TGA analysis, the 4.5% mass loss observed during heating of 3 in the temperature range 200-500 oC under nitrogen atmosphere, relate to the decomposition of the organic part incorporated on MoS2, is consistent with the presence of one functional group for every 49 units of MoS2 (Figure 4.8). Figure 4.9. SEM images for MoS2-based material 3. SEM specimen was prepared by dropcasting a methanol dispersion of MoS2 onto the sample holder and imaged after the solvent was allowed to slowly evaporate. Polygonal overlapping sheets of MoS2 with sizes ranging from hundreds of nanometers to several 89 micrometers in a random distribution were observed (Figure 4.9). Amplification of several areas of the material reveals semi-transparent sheets associated to few layers of MoS2, with regular and linear edges. It should be pointed out that due to the drying process for imaging, re-staking of the MoS2 layers occurs explaining not only the observed deviation from the spectroscopically calculated layer size in solution, but also the difficulty of identifying single-layered MoS2 in the modified material 3. 4.4. Titration experiment Once prepared the positively charged modified MoS2 3 and the negatively charged CDs4, electrostatic attractive interactions between the two species were exploited (Scheme 4.3) for the preparation of CDs/MoS2 complexes. Scheme 4.3. Preparation of the CDs/MoS2 electrostatic complex. The experiment consists of a series of aqueous titration assays. In figure 4.10a are shown the absorption spectra of CDs4 (0.35 mg/ml) for several additions of 3. These spectra are obtained by subtraction of the MoS2 bands, in order to isolate and easily monitor the changes occurring on the CDs absorption band. A progressive red-shift for the absorption of CDs– 4 is observed, namely from 350 to 355 nm after the addition of 320 μL of 3. Moreover, the complex formation between the two species in the ground state is suggested by the presence of a broad isosbestic point at 310 nm. Interestingly, when the neutral MoS2-based derivative 2 was employed for the titration (Figure 4.10b) neither the red-shift nor the isosbestic point formation were observed, proving that electrostatic attractive forces not only promote the CDs/MoS2 complex formation, but also play a role for the effective electronic interaction between the two species. 90 Figure 4.10. UV-Vis absorption spectra of CDs– 4 upon incremental additions of (a) ammonium modified MoS2-based material 3. Inset: Enlargement of the 300-320 nm region where the isosbestic point is developed, (b) MoS2-based material 2. Additional information on the photoinduced electronic interactions between MoS2 and CDs were obtained by steady-state and time-resolved fluorescence spectroscopy analysis. Upon incremental addition of ammonium modified MoS2-based material 3, the CDs emission at 460 nm (λex=370 nm) is progressively quenched (Figure 4.11a). Also when the neutral MoS2-based derivative 2 is added quenching is observed, but with a much lower extent (Figure 4.11b). Nevertheless Stern-Volmer plots provide a useful hint for better understanding the quenching mechanism in the two systems. In fact, while the intensities ratio I0/I (where I0 is the emission intensity of the CDs4 alone and I upon the addition of quencher) increases linearly with the addition of MoS2-based derivative 2, the addition of the ammonium modified MoS2-based material 3 results in a curved Stern-Volmer plot, implying the ocurring of an additional quenching phenomenon. These results suggest that simple collisions are resulting in the dynamic quenching of CDs for the addition of both MoS2-based materials 2 and 3, but only in the case of the latter an additional quenching mechanism is ocurring, which can be reasonably related to the CDs/MoS2 complex formation. 91 Figure 4.11. Photoluminescence titration assays of CDs– 4 (20 μg/mL) upon incremental additions of (a) positively charged MoS2-based material 3, and (b) neutral MoS2-based material 2. Measurements were conducted in water for samples possessing equal absorbance at the excitation wavelength of 370 nm. Figure 4.12. Stern-Volmer plot of CDs 4 I0/I upon incremental additions of (a) positively charged MoS2-based material 3, and (b) neutral MoS2-based material 2. Next, based on the time-correlated-single-photon-counting method, the fluorescence emission decay profiles for CDs– 4 were acquired (Figure 4.13). The analysis of the decay profiles at 460 nm (excitation at 376 nm) for the singlet excited state of CDs– 4 was exclusively monoexponentially fitted with a lifetime of 14.0 ns. Addition of the positively charged MoS2-based material 3 to the negatively charged CDs– 4 resulted in biexponential fitting, giving rise to the identification of two components, namely, one with the same lifetime, attributed to non-interacting CDs and a faster new one with 3.0 92 ns lifetime, corresponding to the fluorescence quenching of the emission intensity of the singlet excited state of CDs within the CDs/MoS2 ensembles. In contrast, the fluorescence decay remained monoexponentially fitted upon addition of neutral MoS2based derivative 2, mantaining the lifetime of intact CDs– 4 and thus excluding in the blank experiment the occurrence of electronic interactions, which instead are proved for the electrostatic complex. Figure 4.13. (a) Decay profiles for CDs– 4 upon incremental additions of positively charged MoS2-based material 3 or (b) neutral MoS2-based material 2. Finally, the electrocatalytic activity of CDs/MoS2 towards the hydrogen evolution reaction (HER) was examined by employing a rotating disc working glassy carbon electrode in a standard three-electrode glass cell at a scan rate of 5 mV/sec in 0.5 M H2SO4. In general, MoS2 are promising materials for HER, based on the overpotential and Tafel slope values they exhibit.12-14 In addition, the aqueous solubility of CDs together with the presence of surface functional groups contribute to draw hydrated protons, thus enhancing proton adsorption capacity.15 Based on the above and considering that hydrogen binds too strongly to S, hence leaving as primary active site for MoS2 the Mo edge, the performance of CDs/MoS2 towards the HER was probed by linear sweep voltammetry. The polarization curve of CDs/MoS2 along with those of individual CDs– 4 and bare glassy carbon electrode for comparison are shown in Figure 4.14a. For a given potential, the cathodic current increased for CDs/MoS2 as compared to that based on individual CDs– 4 and the bare carbon glassy electrode. The evolution of gaseous hydrogen for CDs/MoS2 was visualized as bubbles appearing at currents as small as 0.5 mA/cm2, with enhanced rate at around -0.7 V vs RHE. Since the cathodic 93 current density is proportional to the amount of hydrogen evolved, the latter result highlights the better catalytic activity for CDs/MoS2 and prominent hydrogen evolution behavior exhibiting an onset overpotential near -0.5 V vs RHE, which is lower than the individual CDs– 4 by around 150 mV. The better electrocatalytic activity of CDs/MoS2 is mainly attributed to a synergic effect due to enhanced charge-transfer kinetics owed to the intimate contact between the two species CDs and MoS2 as well as the presence of active sites in MoS2. Next, the linear regions of the Tafel plots (Figure 4.14b) were fit to the Tafel equation, η = B log j + a, where η is the overpotential, j is the current density and B is the Tafel slope, to further characterize the fluent charge transport efficiency and the efficacy of the electrocatalytic reaction. Analysis of the Tafel slope helps to elucidate the possible HER mechanism and define the rate-limiting step. The Tafel slope for CDs/MoS2 ensemble was found to be 22 mV/dec, smaller than the one owed to CDs– 4 by 4 mV/dec. Considering that smaller Tafel slope suggests that for the generation of an equivalent current only a lower overpotential is needed to apply, the electrocatalytic activity of individual CDs– 4 is improved upon realization of the CDs/MoS2 ensemble. The latter improvement in charge transport is attributed to good electrical contact between the two components in the donor-acceptor CDs/MoS2 ensemble, in which charge-transfer phenomena prevail. Moreover, the small Tafel slope of CDs/MoS2 manifests that the electrochemical desorption of adsorbed hydrogen atoms onto the modified electrode to generate hydrogen is the rate-limiting step – see equations 2 and 3 below. Based on the widely applied mechanisms for the HER, initially a proton is adsorbed onto the electrode surface via a reduction process (Volmer adsorption [Eq. (1)]) followed by either direct bonding of the adsorbed hydrogen atom with another proton and electron transfer from the electrode surface (Heyrovsky desorption [Eq. (2)]) or recombination of two hydrogen atoms adsorbed on the electrode surface (Tafel desorption [Eq. (3)]). 94 Figure 4.14. (a) Linear sweep voltammograms for the HER of CDs/MoS2 (black), individual CDs– 4 (red) and bare glassy carbon electrode (dotted). Inset: enlarged region near the onset. (b) Tafel plots for CDs/MoS2 (black) and individual CDs– 4 (red) showing overpotential vs current density. Volmer adsorption: H+ + e–  H•(ads) Eq. (1) Heyrovsky desorption: H•(ads) + H+ + e–  H2 Eq. (2) Tafel desorption: H•(ads) + H•(ads)  H2 Eq. (3) 4.5. Conclusions In this study aqueous stable CDs/MoS2 complexes were developed, profiting of the Coulomb attractive forces for enabling strong electronic interactions between the components. The formation of the complexes was followed by electronic absorption and photoluminescence titration assays, complemented by time-resolved fluorescence emission, proving in this way the occurring of the intra-complex electronic interaction. Significant quenching of the CDs photoluminescence by MoS2 was revealed, prompting to an additional deactivation channel – electron and/or energy transfer – starting from the singlet excited state of CDs within the CDs/MoS2 ensembles. Moreover, the electrocatalytic performance of CDs/MoS2 was evaluated regarding the HER and found improved in comparison with that of the individual CDs species. Without a doubt, such CDs/MoS2 ensembles performing in electron donor-acceptor schemes can be further exploited for managing charge-transfer processes as well as for electrocatalysis and may be useful for advancing the field of energy conversion in a wide range of technological and environmental applications. 95 4.6. References 1. Chhowalla, M.; Shin, H. S.; Eda, G.; Li, L.-J.; Loh, K. P.; Zhang, H., The chemistry of two-dimensional layered transition metal dichalcogenide nanosheets. Nature Chemistry 2013, 5, 263. 2. Wang, Q. H.; Kalantar-Zadeh, K.; Kis, A.; Coleman, J. N.; Strano, M. S., Electronics and optoelectronics of two-dimensional transition metal dichalcogenides. Nature Nanotechnology 2012, 7, 699. 3. Lee, H. S.; Min, S.-W.; Chang, Y.-G.; Park, M. K.; Nam, T.; Kim, H.; Kim, J. H.; Ryu, S.; Im, S., MoS2 Nanosheet Phototransistors with Thickness-Modulated Optical Energy Gap. Nano Letters 2012, 12 (7), 3695-3700. 4. Jeong, H.; Bang, S.; Oh, H. M.; Jeong, H. J.; An, S.-J.; Han, G. H.; Kim, H.; Kim, K. K.; Park, J. C.; Lee, Y. H.; Lerondel, G.; Jeong, M. S., Semiconductor– Insulator–Semiconductor Diode Consisting of Monolayer MoS2, h-BN, and GaN Heterostructure. ACS Nano 2015, 9 (10), 10032-10038. 5. Perkins, F. K.; Friedman, A. L.; Cobas, E.; Campbell, P. M.; Jernigan, G. G.; Jonker, B. T., Chemical Vapor Sensing with Monolayer MoS2. Nano Letters 2013, 13 (2), 668-673. 6. Ross, J. S.; Klement, P.; Jones, A. M.; Ghimire, N. J.; Yan, J.; Mandrus, D. G.; Taniguchi, T.; Watanabe, K.; Kitamura, K.; Yao, W.; Cobden, D. H.; Xu, X., Electrically tunable excitonic light-emitting diodes based on monolayer WSe2 p–n junctions. Nature Nanotechnology 2014, 9, 268. 7. Wang, K.; Wang, J.; Fan, J.; Lotya, M.; O’Neill, A.; Fox, D.; Feng, Y.; Zhang, X.; Jiang, B.; Zhao, Q.; Zhang, H.; Coleman, J. N.; Zhang, L.; Blau, W. J., Ultrafast Saturable Absorption of Two-Dimensional MoS2 Nanosheets. ACS Nano 2013, 7 (10), 9260-9267. 8. Huang, X.; Zeng, Z.; Zhang, H., Metal dichalcogenide nanosheets: preparation, properties and applications. Chemical Society Reviews 2013, 42 (5), 1934-1946. 9. Pumera, M.; Sofer, Z.; Ambrosi, A., Layered transition metal dichalcogenides for electrochemical energy generation and storage. Journal of Materials Chemistry A 2014, 2 (24), 8981-8987. 10. Canton-Vitoria, R.; Sayed-Ahmad-Baraza, Y.; Pelaez-Fernandez, M.; Arenal, R.; Bittencourt, C.; Ewels, C. P.; Tagmatarchis, N., Functionalization of MoS2 with 1,2dithiolanes: toward donor-acceptor nanohybrids for energy conversion. npj 2D Materials and Applications 2017, 1 (1), 13. 11. Li, H.; Zhang, Q.; Yap, C. C. R.; Tay, B. K.; Edwin, T. H. T.; Olivier, A.; Baillargeat, D., From Bulk to Monolayer MoS2: Evolution of Raman Scattering. Advanced Functional Materials 2012, 22 (7), 1385-1390. 12. Wang, D.; Xiao, Y.; Luo, X.; Wu, Z.; Wang, Y.-J.; Fang, B., Swollen Ammoniated MoS2 with 1T/2H Hybrid Phases for High-Rate Electrochemical Energy Storage. ACS Sustainable Chemistry & Engineering 2017, 5 (3), 2509-2515. 13. Chou, S. S.; Sai, N.; Lu, P.; Coker, E. N.; Liu, S.; Artyushkova, K.; Luk, T. S.; Kaehr, B.; Brinker, C. J., Understanding catalysis in a multiphasic two-dimensional transition metal dichalcogenide. Nature communications 2015, 6, 8311-8311. 14. Li, Y.; Wang, H.; Xie, L.; Liang, Y.; Hong, G.; Dai, H., MoS2 Nanoparticles Grown on Graphene: An Advanced Catalyst for the Hydrogen Evolution Reaction. Journal of the American Chemical Society 2011, 133 (19), 7296-7299. 15. Li, H.; Liu, J.; Guo, S.; Zhang, Y.; Huang, H.; Liu, Y.; Kang, Z., Carbon dots from PEG for highly sensitive detection of levodopa. Journal of Materials Chemistry B 2015, 3 (11), 2378-2387. 102 filtrate is washed three times with Et2O. The excess of butylamine is removed under vacuum in a rotary evaporator. Finally the water phase is purified by dialysis in ultrapure water (molecular weight cut-off = 0.5-1 kDa, 3 days). The dry product CDs (0.7g), consisting of a yellow powder, is obtained by freeze-drying. Scheme 5.1a. In the second step 200 mg of CDs were dissolved in 5 ml of MeOH and 25 ml of dichloromethane (DCM) and cooled in ice-bath. In another flask, 300 mg of lipoic acid were dissolved in 5 ml of DCM, cooled in ice-bath and 336 mg of (3Dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride (EDC) coupling agent were added. After 20 minutes, the lipoic acid solution was poured to the CDs one. The reaction was left overnight stirring and the day after 100 mg of lipoic acid and 56 mg of EDC were added. After 3h the reaction was washed one time with water, one time with a NaOH water solution (pH=11) and one time with brine. The DCM was partially removed by vacuum and diluited with ethyl acetate, causing the precipitation of the particles (centrifugation at 3200 r.p.m., 5 min.). The redispersion-precipitation was repeated until the lipoic acid spot in TLC disappeared. The precipitate was dissolved again in MeOH/DCM 1:1 and MgSO4 was added to dry the water traces. f-CDs, in the form of a brown solid, were obtained by vacuum removal of the solvent. Scheme 5.1b. 5.3.2.2 CDs characterization The structure of CDs and f-CDs was determined by 1H NMR spectroscopy (Figure 5.1). All the signals for the as-synthetized CDs are found in the aliphatic region and correspond to the methylene protons of citric acid and ethylenediamine structural components (4.0-2.3 ppm) and butyl units (1.6-0.7 ppm). Thus, the polymer nature of the structure and the occurring of the butylamine functionalization were confirmed. In the modified f-CDs, multiplets appear between 2.6 and 1.5 ppm, corresponding to the protons of the lipoic amide moiety condensed onto the surface of the nanoparticles. 103 Figure 5.1. H1 NMR spectra of CDs (blue) and f-CDs (red). The IR features in the as-produced CDs (Figure 5.2) are assigned as to O-H and N-H stretching vibrations at 3300-3000 cm-1, C-H stretching at 2920 cm-1, C=O stretching of free carboxylic acid units at 1705 cm-1, C=O stretching of amide units at 1640 and 1630 cm-1, and C-O and C-N stretching at 1440-1360 cm-1. In the modified f-CDs, the intensity for the C=O stretching amide band increased. Figure 5.2. ATR-IR spectra of as-produced CDs (blue) and 1,2-dithiolane modified f-CDs (red). 104 Further proof of the functionalization was given by the Kaiser test, where the amount of free amines on CDs decreased significantly upon condensation with lipoic acid, from 1810 to 72 µmol/g. The 1,2-dithiolane functionalization had an impact on the CDs optical properties, resulting in a red-shift from 345 to 370 nm for the f-CDs absorbance and from 450 to 470 nm for the f-CDs emission (Figure 5.3). Figure 5.3. UV-Vis (left) and emission (right, λex=370 nm) spectra of as-produced CDs (blue) and 1,2-dithiolane modified f-CDs (red), obtained in methanol. 5.3.4. Preparation of exfoliated MoS2 and WS2 Bulk TMDs (150-200 mg) were dispersed in chlorosulfonic acid and sonicated for 2 hours at room temperature. The solution was left under stirring during a month, occasionally sonicated for 30 seconds. Afterwards cold water was added to the solution under stirring, drop by drop and extremely carefully. Please notice that the reaction is exothermic and releases gaseous HCl. Next, the mixture was filtrated on a PTFE filter of 0.2 μm pore-size and washed with a good amount of methanol and acetone. The solid compound was added to N-methyl pyrrolidone and sonicated for 1 hour (tip sonication at 30-35% of amplitude (100% of 200 W)). After 3 days the supernatant was taken, filtrated on PTFE filter (0.2 μm pore-size) and washed with a large amount of methanol, acetone and dichloromethane. 5.4. Results and discussion The f-CDs were conjugated to MoS2 and WS2 following the the functionalization methodology for TMDs with 1,2-dithiolanes.39-41 In here, the f-CDs (50 mg) were dissolved in methanol (1 mL). In another flask, exfoliated TMDs (20 mg) were 105 dispersed in DMF (10 mL) by sonication (10 min) and dropped in the CNDs solution. The flask was covered with aluminum foil and the reaction mixture was stirred at 70 °C for 4 days. After that period, it was cooled and filtered through a PTFE membrane (0.2 nm pore size). The solid residue was extensively washed with methanol and dichloromethane to obtain the CD-TMDs (Scheme 5.2). Scheme 5.2. Illustrative preparation of CD-MoS2 and CD-WS2 upon covalent 1,2dithiolane functionalization of exfoliated semiconducting MoS2 and WS2 nanosheets. The as obtained CD-MoS2 and CD-WS2 hybrids were furtherly characterized by IR and Raman spectroscopy, thermogravimetric analysis (TGA) and transmission electron microscopy (TEM). In the ATR-IR spectra of CD-MoS2 and CD-WS2 (Figure 5.4), the presence of CDs was revealed by the bands at 1640 and 1550 cm-1, corresponding to the amide C=O stretching and N-H bending modes, as well as by the sharp bands at 3002800 cm cm-1, due to the C-H stretching. 106 Figure 5.4. ATR-IR spectra of of CD-MoS2 (black) and CD-WS2 (grey). Raman spectroscopy is a useful technique for proving the ocurrence of the covalent functionalization. The spectra of exfoliated MoS2 and WS2 were compared with the ones of the respective CD-MoS2 and CD-WS2 hybrids, obtained upon excitation under onresonance conditions at 633 nm and normalized at the A1g mode at 404 cm-1 (Figure 5.5). In this way, it was found that the intensity of the 2LA(M) band of MoS2, located at 447 cm-1 and associated to disorder and defects,9 decreased after the functionalization of MoS2 with f-CDs, as a consequence of the reduced number of S defects (Figure 5.5a). Moreover, the absence of the characteristic phonon modes of metallic polytype MoS2 so-called J1, J2 and J3 at 150, 225 and 325 cm-1, respectively,10-11 ascertained the semiconducting behavior of MoS2 in the CD-MoS2 hybrid material. Regarding CDWS2, bands due to 2LA(M), E12g, and A1g, upon on-resonance excitation at 514 nm, were found at 350, 354 and 419 cm-1, with the intensity of the 2LA(M) mode decreased by 20% as compared to exfoliated WS2 (Figure 5.5b). Furthermore, for both CD-MoS2 and CD-WS2, the A1g and E12g modes red-shifted by 1-2 cm-1 as compared to the values registered for exfoliated MoS2 and WS2, respectively. The latter is attributed to intrahybrid charge-transfer phenomena developed between f-CDs and the TMDs, in accordance with literature reports.12-13 107 Figure 5.5. Raman spectra normalized at A1g mode for (a) exfoliated MoS2 (blue) and CDMoS2 (black) at λexc 633 nm, and (b) exfoliated WS2 (blue) and CD-WS2 (grey) at λexc 514 nm. Since CDs are highly fluorescent, weak and broad Raman bands attributed to –NC=O, C=O and C-H units (1700-1200 cm-1 and 700-500 cm-1) were observed for both CDMoS2 and CD-WS2 only upon excitation at 1064 nm (Figure 5.6). Figure 5.6. Raman spectra (1064 nm) for CDs (red), CD-MoS2 (black) and CD-WS2 (grey). TGA was employed for evaluating the loading of CDs conjugated onto MoS2 and WS2 in CD-MoS2 and CD-WS2. When heated under nitrogen atmosphere, f-CDs lost the 65% of mass before reaching 500ºC (Figure 5.7). Since MoS2 and WS2 are thermally stable in that temperature range, the observed mass loss at 500 ºC for CD-MoS2 and CD-WS2, 7.5% and 3.0% respectively, is related to the decomposition of f-CDs present 108 in the two hybrids. It should be noticed that tungsten atoms are around two times heavier than molybdenum atoms, thus, for an equivalent load of CDs, the WS2 weight percentage in CD-WS2 is considerably greater than the MoS2 weight percentage in CDMoS2. Taking this into account, the functionalization rates result similar in the two hybrids. Moreover, although this is a relatively small mass loss, it is consistent with the edge functionalization of the limited S vacant sites of MoS2 and WS2.3 Figure 5.7. Thermographs for CDs (red), exfoliated MoS2 (dotted black), exfoliated WS2 (dotted gray), CD-MoS2 (black), and CD-WS2 (grey). The morphology of CD-MoS2 and CD-WS2 hybrids were imaged by HR-TEM. A few drops of a dispersion of the materials in hexane were deposited on the TEM grid and imaged after the solvent was evaporated. Extensive imaging of several different areas and flakes of the CD-MoS2 and CD-WS2 hybrid materials revealed that the size of MoS2 and WS2 is in the order of few hundred nanometres, e.g. around 200-400 nm (Figure 5.8). 109 Figure 5.8. Representative low-magnification HR-STEM-ADF images for CD-MoS2 (left) and CD-WS2 (right). Although mostly oligolayered flakes were observed, most likely due to restacking of the TMDs during the drying process of the sample after depositing it on the TEM grid, the presence of some monolayered ones were also identified. In order to get better insight on CD-MoS2 and CD-WS2, TEM studies complemented with spatially-resolved electron energy loss spectroscopy (EELS) were performed. Figures 5.9a and 5.10a show high angle annular dark field (HAADF) scanning TEM (STEM) micrographs for CD-MoS2 and CD-WS2, respectively. Based on the following spectroscopic/chemical TEM analyses, the bright small objects observed in these images was assigned to f-CDs covalently anchored on TMD flakes. This is confirmed by energy dispersive X-ray spectroscopy (EDS) analyses (Figures 5.9b and 5.10b) and EELS (Figures 5.9c-e and 5.10c-e). Figure 5.9c displays an ADF micrograph of CD-MoS2 and an EELS spectrumimage (SPIM) was recorded in the red marked rectangular area. Three EEL spectra were extracted in the highlighted square regions (Figure 5.9e(i)-(iii)). Each of these three EEL spectra corresponds to the sum of 9 spectra (3x3 probe positions of the SPIM). The S-L2,3 and Mo-M edges are visible in the three spectra and correspond to MoS2.14-15 It is worth mentioning that no MoOx was observed highlighting the high quality and purity of the materials. In addition, C was detected in specific areas, see the presence of the CK edge (Figure 5.9e(ii)-(iii)). This C-K edge, which is superposed to the Mo-M4,5 edge, is associated with the presence of CDs within the CD-MoS2 hybrid. The chemical C map obtained from the analysis of C-K edge (Figure 5.9d) clearly supports this finding.14-15 Indeed, f-CDs are observed not only in the ADF-STEM micrograph (Figure 5.9c) but also in this C-map (Figure 5.9d). Similar assays were performed for CD-WS2 and from TEM analyses (Figures 5.10a-e) the presence of f-CDs attached on WS2 was confirmed. 110 Figure 5.9. Representative HRSTEM-ADF images for (a, d) CD-MoS2. (b) EDS acquired on the squared white area in (a). In the red regions of (c) spectra images of SR-EELS were recorded. (d) Carbon elemental map extracted from the integrated intensity of the C-K edge of the EELS spectrum image recorded in the red area in (c). (e) Three spectra from the sum of nine (3 × 3) EEL spectra extracted from the marked areas of the EELS SPIM of (c). The C-K edge (∼284 eV) is observed in (ii) and (iii) superposed with the Mo-M edge. The S-L2,3 and Mo-M edge of MoS2 are visible in the three spectra ((i)−(iii)). 111 Figure 5.10. Representative HRSTEM-ADF images for (a, d) CD-WS2. (b) EDS acquired on the squared white area in (a). In the red regions of (c) spectra images of SR-EELS were recorded. (d) Carbon elemental maps extracted from the integrated intensity of the C-K edge of the two EELS spectra image recorded in the red area in (c). (e) Three spectra from the sum of 16 (4 × 4) EEL spectra extracted from the EELS SPIM of (c), showing the SL2,3 and C-K (in this case only in (ii) and (iii)) edges. The C-K edge (∼284 eV) is observed in (ii) and (iii). The S-L2,3 edge is visible in the three spectra ((i)−(iii)). The CD-MoS2 and CD-WS2 were analyzed by electronic absorption and fluorescence spectroscopy. The UV-Vis spectra of CD-MoS2 and CD-WS2 (Figure 5.11) confirmed the preservation of the semiconducting monolayered form, characterized by the bands centered at 680, 620, 485, 400 nm, and 645, 535, 475, 420 nm, for MoS2 and WS2 respectively, in addition to continuous absorption throughout the visible region due to the presence of both TMDs and f-CDs (Figure 5.11a). Unfortunately, the strong absorption features of TMDs masked the broad band of f-CDs, appearing at 370 nm (Figure 5.3a), thus impeeding to investigate the electronic interaction between the two materials in the ground state. Nevertheless, clear changes are observed in the excited state, where the strong emission of f-CDs, centred at 470 nm upon excitation at 370 nm, 231 9.3.21. UV/vis spectra of CDs 3a-e Figure 9.104. UV/vis spectra of CDs 3a-e. 232 9.3.22. Excitation and emission spectra of CDs 3a-e CDs 3a Figure 9.105. (left) excitation spectra for different emissions and (right) emission spectra for different excitations of CDs 3a. CDs 3b Figure 9.106. (left) excitation spectra for different emissions and (right) emission spectra for different excitations of CDs 3b. 233 CDs 3c Figure 9.107. (left) excitation spectra for different emissions and (right) emission spectra for different excitations of CDs 3c. CDs 3d Figure 9.108. (left) excitation spectra for different emissions and (right) emission spectra for different excitations of CDs 3d. 234 CDs 3e Figure 9.109. (left) excitation spectra for different emissions and (right) emission spectra for different excitations of CDs 3e. 235 9.3.23. Fitting for the QY calculation of CDs 3a-e Figure 9.110. absorbance vs integrated fluorescence intensity of CDs 3a-e at different concentrations. The slope of the linear fitting was used for calculating the QY. 236 10. ANNEX D 237 10.1. Abstract In this study, sodium borohydride (NaBH4) reactivity was tested towards the CD-MoS2 hybrid prepared in Chapter 5. The addition of NaBH4 to a CD-MoS2 dispersion was reflected in the cleavage of the thiol-molybdenum bond between CDs and MoS2 and therefore in the loss of the covalent functionalization. 10.2. Experimental section For the preparation of the hybrid CD-MoS2 see chapter 5. 10.3. Results and discussion After dispersing 10 mg of CD-MoS2 in DMF by sonication, a centrifugation (5000 r.p.m., 5 minutes) was performed in order to precipitate the material and the supernatant was collected. As expected, no trace of CDs was detected in the supernatant by UV/vis or PL spectroscopy, being CDs firmly attached onto MoS2 by covalent bond. Therefore, the same material was redispersed in DMF and this time few milligrams of NaBH4 were added. The absorption (Figure 10.1a) and fluorescence emission (Figure 10.1b) of the dispersion were measured before and after the NaBH4 addition. Interestingly, after the addition, the emission intensity of the CDs increased considerably, suggesting that the covalent bond with MoS2 was broken and the fluorescence was not affected anymore by its quenching effect. In order to confirm this hypothesis, after the addition of NaBH4 the dispersion was centrifuged as done previously and the supernatant was separated from the precipitate, which was filtered and washed with DMF and water. The IR profile of the dry precipitate does not present anymore the characteristic bands of the CDs between 1700-1600 cm-1 belonging to C=O stretching (figure 1.10c). On the contrary the UV/vis spectrum of the supernatant clearly shows the absorption profile of the CDs (figure 1.10d). These evidences further highlight the key role of the stable covalent bond in the occurrence of the electronic interaction between CDs and MoS2. Additionally, it is shown that a strong reducing agent as NaBH4 is able to break the thiol-molybdenum bond. 238 Figure 10.1. UV/Vis spectra of the CD-MoS2 dispersion before (black) and after (red) NaBH4 addition, (b) fluorescence emission spectra (λexc 370 nm) of the CD-MoS2 dispersion before (black) and after (red) NaBH4 addition, (c) IR spectra of the dry precipitate obtained by centrifugation before (black) and after (red) NaBH4 addition, (d) UV/Vis spectra of the supernatant obtained by centrifugation before (black) and after (red) NaBH4 addition. 239 11. ANNEX E 240 11.1. Abstract In this study, the synthesis of fluorescent CDs is accomplished by microwave irradiation of a solid mixture of citric acid and urea. Filtration and dialysis allowed to separate the raw material by size, obtaining four different fractions. The structural characterization of the fractions was performed by infrared spectroscopy and elemental analysis, furnishing interesting information for the understanding of the nanoparticles growth process. Additionally, differences in the optical behavior were pointed out by absorption spectroscopy and steady state fluorescence spectroscopy characterization. Finally, the full dataset was interpreted under the light of the findings reported in Chapter 2. 11.2. Experimental part 11.2.1. Materials Citric acid monohydrate (99.5%) and urea (99%) were used without further purifications. Dialysis tubes with molecular weight cut-off (MWCO) 0.5-1 KDa were bought from Spectrum Labs. 11.2.2. Characterization techniques  The microwave-assisted reaction was performed in a CEM Discover SP reactor employed in a closed vessel.  Elemental analyses were performed in a Thermo Flash EA 1112 instrument with ∼3 mg of powder samples.  Infrared absorption measurements were performed on powder samples pressed with KBr into pellets with a Bruker Vertex 70 spectrometer.  UV/Vis absorption spectra were recorded on a Shimadzu UV-2401 PC spectrophotometer.  Photoluminescence excitation and emission spectra were recorded on a Horiba Jobin Yvon Fluoromax-P, slits of excitation and emission at 1 mm. All the spectra were recorded at room temperature using 10 mm path-length quartz cuvette. 247 LIST OF SCIENTIFIC CONTRIBUTIONS 248 Publication 1: Supramolecular-enhanced charge-transfer within entangled polyamide chains as origin of the universal blue fluorescence of polymer carbon dots Lorenzo Vallan, Esteban P. Urriolabeitia, Fernando Ruipérez, Jon Mattin Matxain, Ruben Canton-Vitoria, Nikos Tagmatarchis, Ana M. Benito, Wolfgang K. Maser Journal of the American Chemical Society 2018, 140 (40), 12862-12869. Contribution: experimental part (synthesis, most of the characterization), preparation of the manuscript. Publication 2: Electronic Interactions in Illuminated Carbon Dot/MoS2 Ensembles and Electrocatalytic Activity towards Hydrogen Evolution Ruben Canton-Vitoria, Lorenzo Vallan, Esteban Urriolabeitia, Ana M. Benito,Wolfgang K. Maser,and Nikos Tagmatarchis Chemistry – A European Journal 2018, 24 (41), 10468-10474. Contribution: experimental part (synthesis, most of the characterization) and the first article draft. Equal contribution of first and second author. Publication 3: Interfacing transition metal dichalcogenides with carbon nanodots for managing photoinduced energy and charge-transfer processes Lorenzo Vallan, Ruben Canton-Vitoria, Habtom B. Gobeze, Youngwoo Jang, Raul Arenal, Ana M. Benito, Wolfgang K. Maser, Francis D’ Souza, Nikos Tagmatarchis Journal of the American Chemical Society 2018, 140 (41), 13488-13496. Contribution: experimental part (synthesis, part of the characterization) and the first article draft. Equal contribution of first, second and third author. Publication 4: A versatile method for the controllable room-temperature synthesis and in-situ functionalization of fluorescent carbon dots (submitted to Angewandte Chemie International Edition) Lorenzo Vallan, Esteban P. Urriolabeitia, Ana M. Benito, Wolfgang K. Maser Contribution: Synthesis and characterization of the materials (except NMR characterization), preparation of the manuscript.