Hi Everyone, last week one of the fruits of my PhD labour was published in Microchimica Acta. This was a large collaborative work between researchers here in the Nann group at the MacDiarmid Institute as well as researchers at Chalmers University of Technology in Gothenburg, Sweden and the University of South Australia in Adelaide.
We started working on graphene quantum dots after a few publications came out alerting us to this cool fluorescence behaviour the material appeared to have, such as tuneable fluorescence, excitation wavelength-dependent emission, photon upconversion as well as high quantum yields. This seemed too good to be true!
The more we looked into it, the more we started to realise it was too good to be true. Conflicting reports in the literature alerted us to the stark contrast between bottom-up synthesised GQDs using multi-step organic syntheses and their low energy (red) emission and the much higher energy (blue) emission of harsh, top-down, Hummers'-based synthetic methods.
This called into question the very idea of a graphene quantum dot. If it was claimed that a graphene quantum dot was pristine and it was emissive, what could possibly be the origin this emission? Should a truly pristine quantum dot emit?
So we broke the literature down. What was the evidence for a pristine graphene lattice? What we would expect is a very low ratio of oxygen to carbon in XPS analysis, evidence of crystal planes in HR-TEM corresponding to the lattice parameter for graphene or ideally, HAADF TEM showing a perfect graphene lattice (although we know not all groups have access to such facilities). Then Raman spectroscopy showing a very low D:G band integrated intensity ratio - this is a measure of the out-of-plane to in-plane vibrations in the graphene structure, the former being caused by defects in the structure or deviations away from the ideal angles of sp2 bonding.
Another hint at the structure could be the solvent the particles can be suspended in - we expect pristine GQDs to suspend in similar organic solvents to bulk graphene.
Of course there are issues with all of these techniques as a measure of a pristine graphene structure. The major limitation of the three is sample size. Both XPS and Raman will measure a certain spot size which may or may not accurately represent the bulk of your sample. TEM would be the worst technique of the 3 for getting an accurate, statistically relevant description of your entire sample as obtaining HAADF data for every particle would take more time than a PhD.
So what did we do? We developed a milder, exfoliative synthesis for graphene quantum dots and directly compared all of the characterisation to graphene oxide quantum dots synthesised using a harsh Hummers'-based method from the literature. The graphene quantum dots we made could be suspended in mid polarity organic solvents such as THF and DCM - suggesting we were on the right track to making pristine particles. Our XPS data confirmed that we had a very low oxygen content in comparison with the GOQDs and while this was promising, comparing the photoluminescence data with computational studies - namely a very thorough investigation by Sk et al. - we found that our GQDs had much higher energy fluorescence emission than theory predicted. I.e. we had higher energy emission than would be expected of a pristine graphene lattice of the same size (3-4 nm).
Below is a schematic of what we thought the structure was like at this point. Based on the work by Sk et al. and experimental papers as well as our own PL data, the bottom up GQDs had the longest conjugation length and thus lowest energy emission. The top-down GQD using the gentler liquid-phase exfoliation had multiple domains of shorter conjugation lengths leading to green emission. The GOQDs - were a more extreme example where additional defects caused even smaller domains of conjugated carbon and therefore the highest energy emission of all.
To get a better idea of the structure we enlisted the help of a very respected Raman group at the MacDiarmid institute run by Prof. Eric le Ru. They used a number of different Raman experiments to try to resolve the Raman signals over the fluorescence of the GQDs and GOQDs. We found that we could not reliably resolve the Raman signals over the fluorescence background and looking at some of Eric's previous literature, Nile Blue A excited at 633 nm – with a quantum yield of 4% in water - cannot be resolved over its fluorescence using standard techniques. So how could other groups have managed to get Raman spectra? We are still not sure. What is possible though is that given the spot analysis nature of Raman spectroscopy, perhaps other groups have been measuring non-fluorescent portions within the distribution of carbon material produced using these harsh top-down methods.
Finally, what did we learn about the fluorescence? The key differences in optoelectronic properties were:
GQDs have broader excitation (below, blue) and emission (below, red) spectra than GOQDs
GQDs have shorter lifetimes and
GQDs have higher quantum yields
What does this suggest?
It suggests there are a lower proportion of bright states for GOQDs – particularly the longer-lived bright states. We propose that during the harsh chemical synthesis, the bright states similar to those that exist in the GQDs are converted to non-emissive states, leading to lower quantum yields and leaving only the regions with shorter lived fluorescence.
Effect of functional groups
We took the GOQDs in water and resuspended them in less polar, aprotic DMSO and in the process we observe a significant red-shift from 550 to 625 nm. When we resuspended the GQDs in DMSO we observed to spectral shift. This shift to lower energy along with the broadening of the emission spectrum of GOQDs must be largely attributed to the high proportion of oxygen functionalities in the GOQDs – as shown by XPS - that are not as abundant in the GQDs. Red-shifts due to functional groups have been well documented in the literature and there are many reports of pH-dependent emission behaviour of graphene oxide and graphene oxide quantum dots – attributed to the protonation or deprotonation of phenolic and carboxylic acid groups in the structure. In this case, only the GOQDs exhibit a solvent-dependant shift in emission when going from a polar protic solvent (water) to a polar aprotic solvent (DMSO). We therefore postulate that, in an analogous manner to the pH-dependence of fluorescence many groups have reported, the shift here is attributed to the presence or absence of protons rather than the polarity of the solvent.
Our work shows that the origin of the emission in GQDs and GOQDs is a complicated! It involves a combination of edge group emission and emission from conjugated areas within individual particles. Although at first glance, GOQDs appear to have simpler spectroscopic behaviour, this is likely to only be true of bright states and to be related to the introduction of an abundance of dark states during the harsh top-down synthetic method. With this study, we hope to advance the understanding of the structural origin of complex photophysical processes occurring in graphene quantum dots produced using top-down techniques but also to highlight the complexity of structural and spectroscopic data derived from highly heterogeneous materials like these and to open up a discussion on heterogeneity and its effect on interpretation of characterisation data in general.
If you found this work interesting you might also be interested to other papers we've written on heterogeneity and nanomaterials here.
Thanks for reading!