Ground-State Heterogeneity along with Fluorescent Byproducts

Jan 4, 2019 - The integrity of fluorescent carbon dot (FCD) emission deserves its highest appreciation when sample purification is performed with extr...
0 downloads 0 Views 2MB Size
Subscriber access provided by United Arab Emirates University | Libraries Deanship

Spectroscopy and Photochemistry; General Theory

Ground State Heterogeneity along with Fluorescent Byproducts Cause the Excitation-Dependent Fluorescence and TimeDependent Spectral Migration in Citric Acid Derived Carbon Dots Krishna Mishra, Somnath Koley, and Subhadip Ghosh J. Phys. Chem. Lett., Just Accepted Manuscript • DOI: 10.1021/acs.jpclett.8b03803 • Publication Date (Web): 04 Jan 2019 Downloaded from http://pubs.acs.org on January 7, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry Letters

Ground State Heterogeneity along with Fluorescent Byproducts Cause the Excitation-Dependent Fluorescence and Time-Dependent Spectral Migration in Citric Acid Derived Carbon Dots Krishna Mishra, Somnath Koley, Subhadip Ghosh* School of Chemical Sciences, National Institute of Science Education and Research, HBNI, Khurda752050, Odisha, India.

ABSTRACT: Integrity of the fluorescent carbon dot (FCD) emission deserves its highest appreciation when sample purification is performed with extreme care. Several controversial phenomena of FCD fluorescence including excitation dependent emission, spectral migration with time and thereby violation of Kasha-Vavilov rule, those sparked intense debate during recent reports, disappeared when we rigorously purified the as-synthesized FCD sample. Purification was performed by first visual silica column chromatography (observing the emissions under UV-illumination) and subsequently a prolonged membrane dialysis. Most of the surprising phenomena of FCD fluorescence reported earlier apparently arose from a ground state spectral heterogeneity of FCD sample containing a large amount of fluorescent impurities (mostly polymeric or oligomeric in nature). Observation of our ensemble spectroscopic measurements albeit nicely matched with recent reports based on single-particle experiments, but differed largely from other ensemble measurements. Our results reconciled a number of long-standing controversies on FCD emission mostly by emphasizing the urgency of sample purification with more scientific rigor.

1

ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Several attributions of FCD like low cytotoxicity, easy synthesis methods, inexpensive raw materials and excellent spectral properties have tempted researchers for its burgeoning exploration.1-10 FCD particles are quasi-spherical, small ( τfl), it can also appear from ground state spectral heterogeneity (i.e., mixture of fluorescent molecules with different spectral signatures).55-58 Ground state heterogeneity in GO is unlikely as substantiated by its well known structural uniformity, which is not true for FCD. Actual chemical structures and geometry of FCD have not been reported until now. In fact, there have been large ambiguities among the TEM images of FCD particles. For instance, many groups including Strauss et al,59 Khan et al,17 Sharma et al,5 Pan et al40 could not report a meaningful TEM image showing the crystalline structures of FCD, while our group (cf. Figure 1C) including many others like Ehrat et al,50 Ghosh et al,60 Sun et al,61 observed graphitic carbon like fringe structures of FCD particles. TEM image of FCD shows large structural deviations from one particle to other particle (cf. Figure 1C) which would result spectral heterogeneity. In the next part, we will show REES and time dependent spectral migration of FCD fluorescence disappeared when we separated the as synthesized FCD sample into two subsets (BFCD & GFCD) based on their spectral identities. As expected, both REES and spectral migrations were present in the mixed-FCD sample (non-visual 7

ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

column purified).17 Our findings indicate that the REES of FCD fluorescence is neither an intrinsic property (i.e., involvement of multiple emissive sites) nor it appears due to a sluggish solvation around FCD dipole (red edge effect); it arises only from the ground state spectral heterogeneity. Now the obvious question came into our mind, if slow solvation dynamics does not then what is the origin of time dependent spectral migration of FCD fluorescence, observed by Khan et al?17 The magnitude (~post thousand cm-1) and timescale (> ten-nanosecond) of spectral migration reported by Khan et al are unprecedentedly higher than what people generally observed in solvation of organic dye molecule even within a very restricted environment.62-65 Time resolved fluorescence (TRF) profile of a solvated state must associate with rise component(s) during its initial time which would cause the shifting of TRF peak to later time (t=t”) from initial excitation (t=0), unless solvated state is produced at much faster rate than the temporal resolution (~80 ps) of the excitation pulse (Scheme 1A-B).66-70 Population of solvated state grows with time initially, as more number of water molecules are aligned to the excited dipole with time. At the later time (t>t’) of TRF, this solvated state decays due to natural excited state decay process (red curve in Scheme 1B). The rise of population of the solvated state during the initial time is manifested by the appearance of rise component at the early time evolution of TRF profile (red curve in Scheme 1B). A maximum population of solvated state can be observed at later time (t=t’) when formation of solvated state is complete (Scheme 1B). In contrast, TRF profile of un-solvated states must associate with decay component(s) only with maximum population at t=0, i.e., immediately after photo excitation (blue curve in Scheme 1B). Solvated state emits at lower energy (i.e., red emission wavelengths), while un-solvated state emits at higher energy (Scheme 1A). 54, 63-64 A similar argument was made by Wu and coworkers, they observed slow solvent relaxation around GO dipole causing red edge effect, also caused the gradual shifting of TRF peak from zero time (t=0) to some later time (t=t’), as the emission wavelength was shifted toward red side (cf. Figure 4C of ref 54). In bulk water, solvation timescale (τsol~>τfl). In such a scenario, excited fluorophore would emit simultaneously with band-edge’s energy being stabilized through solvation.

Figure 2: Time resolved fluorescence (TRF) of (A) mixed FCD (λex~ 375nm), (B) BFCD (λex~375nm) and (C) GFCD (λex~ 405nm) at extreme blue, extreme red and peak emission wavelengths. In fig B & C all three are identical. Very fast IRF (~80 ps) is shown in black curve. Fast decay curve (pink colored) in figs B (λex~375nm) and C (λex~405nm) depicts the TRF (at emission peak) of citrazinic/succinic acid derivatives, probable impurities in crude FCD. Figure D depicts the plot of average fluorescence lifetime (τavg) as a function of the emission wavelength for all the three FCD samples (BFCD, GFCD and mixed FCD).

This would cause an excitation dependent emission by violating Kasha’s rule and popularly known as red edge effect.54,55,58 Slow solvation causing red edge effect would also cause the observing of initial rise component(s) in TRF profile of red emission (Scheme 1B). However, neither our group (cf. Figure 2B & 2C) nor Khan et al (Figures 3B & 3F of ref 17) observed any such rise component in the TRF profile at red emission, which clearly rule out the notion of red edge effect causing the REES. Absence of rise component is well under or intuitive expectations; in a bulk water, solvation dynamics is ultrafast (10 n spectral migration. (A) Stabilization of highly polar excited state (via solvent-solute interactions) at time scale similar to fluorescence lifetime (τsol~τfl). (B) TRF profiles of the excited state at extreme blue (blue curve; unsolvated), peak (green curve; partially solvated) and extreme red (red curve; solvated) emission wavelengths. (C) Spectral migration of emission with time toward red side due to a gradual stabilization of the excited state energy (Fig A). Spectra in figure C is called time resolved emission spectra (TRES) (D) Emission peak position of TRES as a function of time. Solvation time (τsol) can be obtained from the fitting of this curve, which is proportional to the rise time in TRF of red emission wavelength (red curve in figure B).

Energetically relaxed solvated state emits at red wavelengths with longer fluorescence lifetime as compared to that from un-solvated state. It is worth mentioning here, many groups including Khan et al17 reported the strong dependency of fluorescence lifetime (τfl) of FCD on emission 10

ACS Paragon Plus Environment

Page 11 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry Letters

wavelength, which we believe not due to a slow solvation dynamics, rather it appeared from the heterogeneity of their sample. Our mixed-FCD sample showed a strong emission wavelength dependent fluorescence lifetime (cf. Figure 2D), which disappeared in BFCD and GFCD. If emission wavelength dependent lifetime arose due to slow solvation, it would have been also present in BFCD and GFCD. Slow solvation dynamics (or red edge effect) causes the formation of partially solvated energy states with sharp emission maxima. However, when their contributions are added in the actual steady state emission spectrum, it appears much broader. This spectral broadening is known as inhomogeneous broadening.44 Changing the excitation wavelength would cause a change in relative population of these emissive states. The net result would be an excitation dependent emission and a large spectral migration of TRES (cf. Scheme 1C & 1D).44 The excitation dependent emission and spectral migration can also be found from a mixture of fluorescent particles with different spectral signatures, which need to be duly verified for FCD sample before asserting red edge effect causing the REES.17,44 Figures 3A-C depict the steady state emission spectra of mixed-FCD, BFCD, and GFCD at different excitation wavelengths (λex). The FWHM of the mixed-FCD sample is found always higher than the FWHM of BFCD or GFCD. Large spectral broadening of mixed-FCD sample, which Khan et al wrongly assumed as a result of an inhomogeneous boarding, was actually arose due to the presence of various FCD subsets within their non visual column separated FCD sample (cf. Figure 3A).17 We excited the our FCD samples at all possible wavelengths, starting from extreme blue to the extreme red with respect to their primary absorption peak positions and took the emission spectra (Figure S2). Steady state emission peak of BFCD was found independent to the excitation wavelength, while emission of GFCD shifted slightly with excitation wavelength (Figure 3D). This is in contrast to the mixed-FCD, where a large shifting (> 70 nm) of emission peak was observed with changing the excitation position (cf. Figure 3). Khan et al ascribed this shift is due to a red edge effect which breaks the Kasha’s rule.17 Combining the observation of non-REES phenomenon from pure FCDs and not obtaining any rise component in TRF substantially refuting the earlier claim that believed red edge effect causing the REES for FCD fluorescence.17 A careful examination of Figure 3D reveals the shifting of emission peak position of mixed FCD is not continuous, rather a single step transition; from a BFCD like emission to GFCD like emission as excitation wavelength increases. If the excitation dependent emission occurred due to the red edge effect, one would have observed a continuous shift of emission peak 11

ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 26

instead of a step-wise feature; since solvation dynamics produces continuum of narrowly spaced energy states.63-64 A similar observation of non-REES phenomenon of FCD fluorescence was earlier reported by Ghosh et al using single molecule spectroscopic measurements.60 Their study although showed the emission peak and τfl of a single FCD particle were excitation independent, but their values changed from one particle to other particle.60

Figure 3: Steady state fluorescence spectra of mixed FCD (A), BFCD (B) and GFCD (C) at different excitation wavelengths. Figure (D) shows emission peak as a function of excitation wavelength for all the three samples. Inset of Figure A shows the same spectra after normalizing with peak intensity. A slight REES in GFCD indicating this batch contains two closely emitting subsets; both are emitting at green wavelengths and very difficult to further column separation looking at their emission colors. Ground state heterogeneity can easily mislead the ensemble measurements but not the single

molecule measurements. In the later technique, spectral property from a single particle is measured and if the excitation dependent emission is intrinsic to FCD, one would have observed the same phenomenon from a single FCD particle also.60 Time resolved emission spectra (TRES), which represent the emission spectra at different time after photo-excitation to the FCD, are constructed following the recipe of Fleming and coworker (See SI for further details).71-77 Migration of emission energy to red wavelengths with 12

ACS Paragon Plus Environment

Page 13 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry Letters

time is frequently observed when solvation gradually stabilizes the excited state energy of the emitting dipole (Scheme 1A & 1C). Spectral migration can only be detected when the solvation timescale (τsol) is much slower than the IRF (~80 ps) of excitation lamp. In bulk water spectral migration occurs at ultrafast timescale (~1ps), which can’t be observed using our setup. However, observing a significant spectral migration (>1500 cm-1) at post ten nanosecond timescale from mixed-FCD (non visual column purified) in bulk water is indeed surprising, observed by our group and Khan et al (cf. Figure 4A & 4D).17 Khan et al argued this shift is a result of red edge effect (τsol>τfl),17 but surprisingly not supported by observing an initial rise component in the TRF of FCD at red emission wavelengths, neither by our group (cf. Figure 2A) nor by khan et al (cf. Figure 3B and 3F of ref 17). Interestingly, like excitation dependent emission, the spectral migration also disappeared when we visually column separated the FCD sample to two spectrally homogeneous subsets (BFCD & GFCD). Our study substantially refuting the earlier notion of red edge effect (τsol>>τfl)17 causing the spectral migration of FCD emission.51-55,65-69

Figure 4: TRES of mixed FCD (A) show the spectral migration toward low energy side with time, whereas in BFCD (B) and GFCD (C) no shift was observed. Inset shows the same TRES after area normalization (TRANES). A clear iso-emissive point is obtained in TRANES of mixed FCD only (insets of Figure A), which is further zoomed for better clarity. (D) Emission peak position of TRES as a function of time for all the three samples.

13

ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Through Scheme 1, we have shown how solvation dynamics causes the spectral migration of TRES. Now using Scheme S1, we will show how ground state heterogeneity may also cause the similar spectral migration but in that case a rise component in TRF cannot be observed. Correlating with our mixed FCD sample, in scheme S1 we assumed a mixed fluorophore sample consist of two fluorophores emitting at different positions and having different lifetimes (like our mixed FCD). Contrasting to our FCD samples, in the scheme S1, we assumed blue emitting fluorophore has longer τfl than the green emitting fluorophore. Qualitatively scheme S1 will correlate with our mixed FCD sample but here spectral migration will be in opposite direction. Immediately after photo-excitation (i.e, at t=0), both the fluorophores would exhibit maximum fluorescence intensity. When their contributions are added to obtain the sum spectrum, we would observe the peak position of the sum spectrum (at t=0) somewhere in between the emission peaks of the two emitting species (Scheme S1). At a longer time (i.e., at t=t1; 0< t1