Carbon dots, otherwise known as C-dots, are carbon nanoparticles which have diameters of less than 10 nm. Carbon dots were discovered by chance in 2004 by Xu, et al1 during their research into different ways of purifying carbon nanotubes.
Ever since, carbon dots have been explored by the Materials Science Community as a potential alternative to semiconductor quantum dots. Thanks to their low toxicity, much of this research is focussed on their potential in biological applications, as well as in solar cells, photocatalysis, bioimaging and drug delivery.
The composition of C-dots are a mixture of sp2 and sp3-hybridized carbon atoms and functional groups like –COOH, -NH2, and –OH. C-dots can be synthesised by starting from sp2 carbon materials such as graphite- otherwise known as “top down” or from carbonization of small molecules, or “bottom up” (Figure 1).
Though the “bottom up” syntheses enable larger-scale production, they produce very inhomogeneous C-dots. Recent exploration of synthesis of C-dots has an environmental focus, with carbon dots being produced from sucrose, waste oil, or vegetables. 2 This carbonization of organic matter is achieved by heating to temperatures over 200 °C or through treatment with acids.
Figure 1. The “Top-down” and “bottom-up” approaches in the synthesis of C-dots
Many applications of carbon dots utilize their photoluminescence (PL). Typically, C-dots present good PL quantum yields and have a characteristic shift in their emission spectrum with excitation wavelength: The PL spectrum shifts towards the red thanks to longer excitation wavelengths; an apparent break with Kasha’s rule. An Excitation-Emission Map (EEM) of the sample’s photoluminescence can easily identify this behavior.
To explain the wavelength-dependent emission spectrum of –C-dots, several hypotheses have been proposed. One such hypothesis is the effect of carbon dot size, similar to that of semiconductor quantum dots.3 Quantum confinement takes place as the carbon particle size decreases, meaning that the particle is smaller than the de Broglie wavelength of the dot’s electron, which creates a deviation from bulk properties.
As a result of this confinement, a quantization of the energy is created into discrete levels in the conduction and valence bands. This means that the C-dot can now be understood as a “virtual atom”. The radius of the particle determines the emission energy, so that as the particle decreases in size, both the emission spectrum and the excitation shift to shorter wavelengths.
Different subsets get preferentially excited in a distribution of C-dots with varying sizes, as the excitation wavelength varies. The emission spectrum shifts with excitation wavelength as C-dots of different size are excited.
Despite reported size-dependent PL of C-dots, it is not the only factor with the potential to affect the emission spectrum and it is difficult in practice to obtain a distribution which has a narrow PL spectrum. Another theory to potentially explain the PL behavior of carbon dots is the existence of different emissive sites on the dot’s surface.4 These emissive sites are related to different defects on the dot’s surface, which depending on the wavelength used, are selectively excited.
The mechanism of PL in C-dots remains largely uncertain, despite strong efforts of research in this field. The lack of information is partly due to the dearth strategies through which to selectively produce just one type of C-dot structure. New synthetic approaches are needed in order to control the degree of crystallinity and functionalisation of C-dots.
This article characterizes an inhomogeneous distribution of carbon dots produced by the thermal treatment of milk. In this case, heating up milk in air for 5 hours (until it reaches 220 °C) achieves carbonization, and the photoluminescence spectra and lifetimes of the resulting C-dots are examined in an FLS1000 Photoluminescence Spectrometer.
Whole cow’s milk was diluted 1:1 in water and heated up for 2 hours in a conventional oven at 220 °C. The black residue which resulted was collected and extracted with methanol and dichloromethane so that the high and intermediate polarity fractions of C-dots might be obtained.5
C-dot solutions were put in quartz cuvettes and measured in an FLS1000 Photoluminescence Spectrometer equipped with double monochromators, a 450 W Xe lamp, Time-Correlated Single Photon Counting (TCSPC) capability, standard cuvette holder, and standard PMT-900 detector. EEM, or the excitation-emission maps were acquired automatically with the standard option available in the Fluoracle software.
Figures 2 and 3 present the photoluminescence EEMs of C-dots extracted with dichloromethane and methanol respectively. Figure 2 presents characteristic C-dot emission in the range of 400 nm – 600 nm as well as a series of narrow UV bands when exciting at 300 nm – 350 nm (a). This band structure could be due to polyaromatic hydrocarbons, which may be present in heated milk.6
A close-up of C-dot emission (b) demonstrates a very broad spectrum with a clear shift of its peak as the excitation wavelength moves towards the red.
Figure 3 shows an EEM of the methanol fraction of carbon dots. An interesting point to note is that the narrow band structure is not present in the UV region (a) indicating that those compounds are not soluble in methanol. However, the C-dot region (b) is very similar to the dichloromethane phase.
Figure 2. EEM of C-dots extracted with dichloromethane from heated milk: (a) 3D map, (b) 2D map of C-dot region (excitation wavelength shift indicated in the graph). Measurement conditions: Δλex = 3.00 nm, Δλem = 2.00 nm, λex step = 5.00 nm, λem step = 2.00 nm, dwell time = 0.5 s/step.
Figure 3. EEM of C-dots extracted with methanol from heated milk: (a) 3D map, (b) 2D map of C-dot region (excitation wavelength shift indicated in the graph). Measurement conditions: Δλex = 3.00 nm, Δλem = 2.00 nm, λex step = 5.00 nm, λem step = 2.00 nm, dwell time = 0.7 s/step.
Time-resolved PL measurements can provide further information on carbon dots’ luminescence mechanism. A Time-Resolved Emission Spectrum (otherwise known as a TRES) is a sequence of PL lifetime measurements with varying emission wavelength. This provides the spectral shift in emission as a function of time, after the excitation pulse.
Time Correlated Single Photon Counting acquired TRES maps of both samples, as shown in Figures 4 and 5. Both figures corroborate a clear increase in PL lifetime at longer wavelengths; indicating that the emission spectrum shifts to the red with time, as demonstrated in the insets.
Due to the timescale of the process, this is unlikely to be a solvent relaxation effect (it is typically ps rather than ns). The size distribution of the C-dots could have caused the red shift however, existing literature reports indicate the opposite trend, with larger C-dots emitting in the red possess shorter PL lifetimes.7 As a result, the most feasible explanation is the existence of different emissive sites, which possess varying energies and PL lifetimes.
Figure 4. TRES of C-dots extracted with dichloromethane from heated milk, PL intensity colour code in the graph. The red and blue lines show the PL decays at 470 nm and 430 nm emission, respectively. The inset displays peak-normalised emission spectra integrated at different times in the decay (colour code in the graph). Measurement conditions: λex = 405 nm, Δλem = 5.00 nm, λem step = 10.00 nm, time resolution = 49 ps/channel, acquisition time = 5 minutes/decay.
Figure 5. TRES of C-dots extracted with methanol from heated milk, PL intensity colour code in the graph. The red and blue lines indicate the PL decays at 490 nm and 430 nm emission, respectively. The inset shows peak-normalised emission spectra integrated at different times in the decay (colour code in the graph). Measurement conditions: λex = 405 nm, Δλem = 5.00 nm, λem step = 10.00 nm, time resolution = 49 ps/channel, acquisition time = 5 minutes/decay.
The results indicate that thermal treatment of milk is an environmentally friendly and widely available method by which to produce carbon dots. Using the EEM available in the software, the C-dots are easily characterised in an FLS1000 photoluminescence spectrometer. Time-resolved emission spectra using TCSPC offer information about the PL mechanism and further corroborate the diverse character of the C-dots studied.
- X. Xu, R. Ray, Y. Gu, H. J. Ploehn, L. Gearherat, K. Raker and W. A. Scrivens, J. Am. Chem. Soc. 126, 12736-12737 (2004)
- R. Das, R. Bandyopadhyay, P. Pramanik, Materials Today Chemistry 8, 96-109 (2018)
- Edinburgh Instruments, Photophysical Characterisation of Perovskite Quantum Dots (2018)
- A. Sharma, T. Gadly, A. Gupta, A. Ballal, S. K. Ghosh and M. Kumbhakar, J. Phys. Chem. Lett. 7, 3695 – 3752 (2016)
- S. Han, H. Zhang, J. Zhang, Y. Xie, L. Liu, H. Wang, X. Li, W. Liu and Y. Tang, RSC Adv. 4, 58084-58089 (2014)
- C. Naccari, M. Cristani, F. Giofrè, M. Ferrante, L. Siracusa, D. Trombetta, Food Research International 44, 716-724 (2011)
- K. Hola, et al, Carbon 70, 279-286 (2014)
This information has been sourced, reviewed and adapted from materials provided by Edinburgh Instruments.
For more information on this source, please visit Edinburgh Instruments.