Environmentally Friendly Nitrogen-Doped Carbon Quantum Dots for Next Generation Solar Cells

Shine on you crazy carbon! In this work, nitrogen-doped carbon quantum dots (N-CQDs) are synthesized using a simple custom atmospheric pressure microplasma. The method is facile, rapid, and environmentally friendly and the N-CQDs can be produced in a few minutes with no need for high temperature, complicated chemical techniques, or surface ligands. The N-CQDs are formed using molecular precursors and can be produced in different solvent mixtures. Material characterization techniques show a high degree of nitrogen doping on the QD surface with the amount of nitrogen depending on initial reaction conditions. The N-CQDs show interesting quantum confined optical properties that depend on the amount of nitrogen incorporation. Importantly, the band energy structure of the N-CQDs is elucidated and they are incorporated into a photovoltaic device as the photoactive layer achieving an extraordinary open-circuit voltage of 1.8 V and a power conversion efficiency of 0.8% (champion device), amongst the highest reported to date for group IV and carbon based quantum dots.

of the water was adjusted to maintain a constant discharge voltage of 1.3 kV.The reaction was paused and the solution stirred every 10 minutes.The reactions were carried out for 30 minutes in total to give a yield of 1 mg/mL of N-CQDs in water.

Fabrication of solar cells:
Substrates from VisionTek Systems Ltd. were 38 mm in length, 13 mm in width and had a thickness of 0.7 mm.They were fabricated using OLED-quality polished glass and included a patterned strip of indium-doped tin oxide (15 Ω/sq, 150 nm thick, and 2 mm wide) running lengthwise down the middle of the substrate.Before fabrication of devices, substrates were cleaned by successive sonication in propan-2-ol (IPA), immersion in boiling acetone, sonication in fresh IPA, and sonication in DI water for 10 minutes each and finally dried with a nitrogen gun.
A compact TiO 2 hole blocking layer was formed using a sol-gel method.Briefly, titanium isopropoxide (1.56 mL) was added to a clean beaker containing triethanolamine (0.394 g) and ethanol (18 mL).The mixture was stirred on a hotplate at 40 °C for 2 hours.The solution was then left to rest at room temperature overnight.About 100 μL of this solution was then spuncast onto the pre-patterned ITO coated substrates at 4000 revolutions per minute (RPM) for 30 s.The substrate was heated on a hotplate at 100 °C for 30 min and then in an oven at 350 °C overnight to promote crystallization of the film.This process formed a 40 nm compact layer of anatase TiO 2 .
Subsequently, N-CQDs were deposited by spray coating using an atomizer (Sonozap Ultra Sonic Atomizer) operating at a power of 5 W. 5 mL of an N-CQD solution was loaded into a syringe, connected to a pump and flowed through the atomizer at a rate of 0.15 mL/min.The atomizer tip was 8 cm away from the substrate.The substrate was continuously heated on a hotplate at 170 °C throughout deposition and left to dry for 10 minutes after deposition.Gold contacts were deposited orthogonal to the indium-doped tin oxide (ITO) strip, using plasma assisted magnetron sputtering.The plasma was produced at 317 V and 0.15 A. Deposition was carried out for 50 minutes to give a gold film with a thickness of ca.340 nm.The active area of each cell was 4 mm 2 .

Characterization:
Transmission electron microscopy images and selective area electron diffraction patterns were acquired using a high resolution JEOL JEM-2100F field emission electron microscope, and Gatan DualVision 600 Charge-Coupled Device (CCD), operating at an accelerating voltage of 200 keV.TEM samples were prepared by depositing a 40 μL aliquot of the QD dispersion onto a holey carbon-coated copper grid (300 mesh, #S147-3, Agar Scientific), which was allowed to evaporate under ambient conditions.Data for size distribution histograms was acquired by analysis of TEM images of exactly 200 QDs located at different regions of the grid.QD diameter was determined by manual inspection of the digital images.
X-ray diffraction measurements were performed on a Bruker AXS D8Discover instrument using monochromatic Cu Kα X-ray radiation at an accelerating voltage of 40 kV and current of 40 mA.Scans were performed in 0.02 increments and at a scan speed of 20 s per step to give a total acquisition time of 20 hours.Samples were prepared by spray coating colloids with the atomizer on glass substrates.X-ray photoelectron spectroscopy measurements were carried out using a Kratos Axis Ultra DLD photoelectron spectrometer at ≈ 10 -9 mbar base pressure.The narrow scan spectra were obtained under high vacuum conditions using a monochromatic Al Kα X-ray radiation at 15 kV and 10 mA with an analyzer pass energy of 20 eV.The N-CQD dispersion was deposited using the atomizer on either plane glass or ITO coated glass substrates.All spectra were acquired at room temperature and binding energies were referenced to the C 1s line at 285 eV.All spectra were corrected using a linear background fitting.
Ultraviolet photoelectron spectroscopy was carried out with a Kratos Axis Ultra DLD spectrometer and measurements were performed at < 2 × 10 -8 mbar base pressure with He 1 α (h = 21.22 eV) as the excitation source, a dwell time of 0.1 s with 40 sweeps performed.Samples were prepared by spray coating the N-CQDs on ITO coated glass.
Fourier transform infrared spectra were recorded on a Thermo Scientific Nicolet iS5 spectrometer equipped with an iD5 attenuated total reflection (ATR) accessory.Spectra were recorded on colloids that were spray coated using the atomizer onto glass substrates.Spectra were recorded with a resolution of 2 cm -1 and 32 scans were averaged.
Kelvin probe measurements (SKP Kelvin Probe Version Delta 5+, Version 5.05 from KP Technology Ltd., U.K.) were performed to determine the Fermi level of the N-CQDs.This technique measures the contact potential difference (CPD) between the sample surface and a vibrating tip.The work function ( ) of the N-CQD film can be measured in comparison to a  reference material, in this case gold.A thick film was deposited from colloids that were spraycoated onto ITO-coated glass substrates and the ITO layer was connected to the ground during measurements.All measurements were performed at ambient conditions, the probe was scanned across a 100 µm x 100 µm area, taking measurements at 25 different points in the scanned area, and relative changes in the CPD were recorded.The work function ( ) is then calculated using  the equation; where is the work function of Au (5100 meV relative to the vacuum level), is the CPD between the tip and the Au surface while is the CPD between the tip and the sample surface.An average value and   corresponding standard deviation were calculated from the 25 measurements for each sample.
Ambient photoemission spectroscopy (APS) was carried out with a similar instrument with a module containing a high intensity deuterium source coupled with a motorized grating monochromator.The sample is illuminated with a 4-5 mm diameter light spot derived from the tuneable monochromated D2 lamp.Nitrogen gas is used to suppress the production of ozone in the DUV spectrometer and the energy range of incident photons is typically 3-7 eV.The raw photoemission data are corrected for detector offset; intensity normalized, then processed by either a square or cube root power law.
UV-Vis absorption spectra were recorded using a PerkinElmer Lambda 650 S spectrophotometer equipped with a 150 mm integrating sphere.Photoluminescence spectra were recorded using an Agilent Cary Eclipse spectrophotometer.Spectra for both UV-Vis and PL were recorded at room temperature using quartz cuvettes (1 cm) and corrected for the solvent absorption or emission.Data for the Tauc plots were acquired as follows.The transmission ( )  of the colloidal samples was obtained in the normal way in the sampling compartment in front of the integrating sphere.This does not account for scattering or reflectance of the sample.The transmission of the samples was also attained inside the integrating sphere ( ).This considers sample reflectance and scattering ( ).The absorption coefficient ( ) was then  +  +   determined using the following formula: Where is the transmission taken inside the integrating sphere ( ), is the path length (spectra were recorded on colloids in a 1 cm cuvette), and is the transmission taken in  the normal fashion in the front compartment which neglects reflection and scattering.ℎ 1/2 was plotted as a function of energy, and the bandgap determined from the intersection with the x-axis of a linear fit to the data.
For absolute quantum yield measurements, an integration sphere attached to a Horiba Jobin Yvon fluoromax-4 spectrometer was used to collect the PL of the N-CQD film.For excitation, a Xe lamp with a double monochromator was used, and the PL was detected by a chargecoupled detector (CCD) mounted on a spectrograph via coupled ultraviolet-grade optical fiber.
The excitation wavelength was selected through the monochromator.The emission spectra from the sample (N-CQDs on quartz) and the reference (quartz only) were measured, and the number of emitted photons was then calculated from spectral integration.The number of absorbed photons was calculated using reduction of the excitation spectrum and comparing the sample and reference.The absolute QY is obtained as the ratio of the number of emitted photons to the number of absorbed photons.Measurements were carried out with an excitation wavelength of 420 nm (3 eV), performed in triplicates, and the average value was reported.
Electrical characterization of the solar cell devices was performed using a Keithley 2400 SourceMeter and running Tracer 2 software.Cells were 0.04 cm 2 in size and illuminated at 1 sun, AM 1.5 (1000 W/cm 2 ) with a LOT QuantumDesign solar simulator light source.The J-V characteristics of the solar cells were evaluated at room temperature in air.In all cases, the irradiance was calibrated using a standard 4 cm 2 Si solar cell.Because conditions at the anode are more complicated and different non-electrochemical reactions can occur, the microplasma anode provides unique circumstances for preparing N-CQDs, that are different to other reported plasma methods.When the reaction was carried out with a platinum electrode replacing the plasma anode, the color of the solution did not change meaning that no N-CQDs formed, it was not a simple electrochemical reaction, and that the presence of the microplasma was necessary.The suggested formation mechanism is the plasma enhanced condensation reaction between OH containing carbon precursor and nitrogen precursor containing NH 2 groups. 1,2The carbon molecules can self-assemble and condensation reactions occur to form an extended carbon framework.The N atoms enter the QD framework by forming a pyrrolic structure (see FTIR and XPS discussion in the manuscript) through further intramolecular dehydroxylation between adjacent carboxyl and amide groups.In particular, the decrease of the C=O peak in the FTIR upon increasing the amount of nitrogen, suggests the consumption of -COOH groups due to the dehydrolysis reaction discussed above, while the increase of the C=N peak implies the successful N doping with formation of pyrrolic ring-like structures.The changing oxygen and nitrogen XPS atomic percentages, as well as the presence of pyrrolic N further support this formation mechanism.The microplasma as the anode significantly accelerates the rate of this reaction, allowing for the low temperature, rapid synthesis of highly luminescent N-CQDs.

Results and discussion
Solutions with a fixed amount of water and CA, but varying amounts of EDA were treated with the microplasma.When no EDA was present, only a very minor color change was observed from colorless to a very pale straw yellow color, Fig. ESI2 (   From the data shown in Fig. 4 we can calculate the bandgap via a complimentary method.In the Tauc plot, is plotted against energy ( ), where the factor represents the nature of the optical transition.The absorption coefficient ( ) can then be expressed as: Where is a constant, is the energy of photons, and is band gap.From [1], we then have:

Table ESI1 .
Summary of the different types of photovoltaic cells in which GQDs and CQDs are utilized and their performance characteristics Graphene Quantum Dots, b) DSSC: Dye Sensitized Solar Cell, c) PSC: Polymer Solar Cell, d) CQD: Carbon Quantum Dots, e) NRSC: Nanorod Solar Cell, f) NTSC: Nanotube Solar Cell, g) NRSC: Nanorod Solar Cell, h) QDSC: Quantum Dot Solar Cell, i) SMSC: Small Molecule Solar Cell, j) S-CQD: Sulphur doped Carbon Quantum Dots