Quenching of Single-Walled Carbon Nanotube Fluorescence by

Apr 13, 2017 - Because these different structural forms show distinct physical and chemical properties, the sorting of SWCNT mixtures into pure (n,m) ...
17 downloads 11 Views 1MB Size
Download PDF
Letter pubs.acs.org/JPCL

Quenching of Single-Walled Carbon Nanotube Fluorescence by Dissolved Oxygen Reveals Selective Single-Stranded DNA Affinities Yu Zheng,† Sergei M. Bachilo,† and R. Bruce Weisman*,†,‡ †

Department of Chemistry and the Smalley-Curl Institute and ‡Department of Materials Science and NanoEngineering, Rice University, 6100 Main Street, Houston, Texas 77005, United States S Supporting Information *

ABSTRACT: The selective interactions between short oligomers of single-stranded DNA (ssDNA) and specific structures of single-walled carbon nanotubes have been exploited in powerful methods for nanotube sorting. We report here that nanotubes coated with ssDNA also display selective interactions through the selective quenching of nanotube fluorescence by dissolved oxygen. In aqueous solutions equilibrated under 1 atm of O2, emission intensity from semiconducting nanotubes is reduced by between 9 and 40%, varying with the combination of ssDNA sequence and nanotube structure. This quenching reverses promptly and completely on the removal of dissolved O2 and may be due to physisorption on nanotube surfaces. Fluorescence quenching offers a simple, nondestructive approach for studying the structure-selective interactions of ssDNA with singlewalled carbon nanotubes and identifying recognition sequences.

S

ingle-walled carbon nanotubes (SWCNTs) are among the most widely studied artificial nanomaterials. Samples of asproduced SWCNTs contain many discrete structural forms, each of which is labeled by a pair of integers, (n,m), and has a well-defined diameter and roll-up angle. Because these different structural forms show distinct physical and chemical properties, the sorting of SWCNT mixtures into pure (n,m) fractions has been a central goal for enabling basic research progress and developing advanced applications. A remarkably effective method for such sorting is based on structure-selective wrapping interactions of short ssDNA oligomers with SWCNTs. Zheng and coworkers have explored these affinities by suspending SWCNT mixtures with various ssDNA oligos and analyzing the results of chromatographic or aqueous twophase partitioning separation processes.1−3 They have thereby identified ssDNA sequences that have special interactions with certain (n,m) SWCNT structures. Here we report that those structure-selective interactions also cause easily measurable differences in the quenching of SWCNT fluorescence by dissolved O2. This phenomenon allows specific affinities to be readily uncovered and studied through quick and nondestructive optical measurements. The oxygen quenching phenomenon is shown in Figure 1, which plots emission spectra measured from a sample of HiPco SWCNTs dispersed in an aqueous solution of the (GT)20 oligo. Emission from the air-saturated sample is significantly intensified by purging with argon to displace dissolved oxygen. Subsequent purging with pure O2 drops the emission sharply to below the level of the initial air-saturated solution. Finally, resaturation with air precisely restores the initial emission spectrum (to within the width of the lines used in the Figure). This effect is unrelated to the permanent spectral changes induced by covalent oxygen atom doping of SWCNTs.4 We attribute the newly found reversible changes to encounters between O2 and nanotube excitons, acting through a © XXXX American Chemical Society

Figure 1. Fluorescence spectra of SWCNTs suspended by (GT)20 ssDNA. The thick black curve shows the initial spectrum of the airsaturated sample. After purging with argon, the green curve was measured. Subsequent purging with oxygen gave the partly quenched emission shown as the red trace. A final purge with air gave the dashed yellow spectrum, which matches the initial (black curve) spectrum to within the width of the traces in the Figure.

mechanism different from the redox reactions first described by Zheng and Diner.5 In those charge-transfer processes, larger diameter SWCNTs were most easily oxidized, whereas Figure 1 clearly shows that changes in the dissolved oxygen concentration most strongly affect smaller diameter SWCNTs, which emit at shorter wavelengths. The fluorescence quenching found here is also relatively insensitive to changes in sample pH and proceeds too quickly to resolve with our current apparatus (see Figures S1 and S2). Received: March 9, 2017 Accepted: April 13, 2017 Published: April 13, 2017 1952

DOI: 10.1021/acs.jpclett.7b00583 J. Phys. Chem. Lett. 2017, 8, 1952−1955

Letter

The Journal of Physical Chemistry Letters

Figure 2. Relative emission intensity under Ar as compared with O2 for (a) (GT)10 ssDNA-SWCNTs, (b) (ATT)4 ssDNA-SWCNTs, and (c) SDSSWCNTs at SDS concentrations of 0.05 and 0.1%.

To explore the (n,m) dependence of fluorescence quenching by dissolved oxygen, we compared the emission spectra of samples when purged with Ar at 1 atm and when purged with O2 at 1 atm. A chemical buffer (NaH2PO4/Na2HPO4) kept the pH stable at 7.4. By dividing the first spectrum by the second, we obtain a quenching spectrum in which the ratio equals 1 if there is no oxygen-induced quenching and assumes larger values as quenching increases. Figure 2a shows an example of such a quenching spectrum for a sample dispersed in (GT)10 ssDNA. The maximum quenching factor of ca. 2.5 near 1050 nm is assigned to (7,5) SWCNTs; wavelengths characteristic of other (n,m) species show smaller quenching factors in this ssDNA coating. The quenching factors clearly depend on ssDNA sequence as well as nanotube (n,m) structure. This can be seen from the quenching spectrum in Figure 2b, which was measured for a sample suspended in (ATT)4. Here the quenching factor at the (7,5) wavelength is a local minimum instead of a maximum as it was in (GT)10. The dependence of oxygen quenching factor on (n,m) and DNA oligo structures suggests that a key factor may be how completely the ssDNA covers the nanotube surface. To investigate this hypothesis, we measured SWCNT quenching by dissolved oxygen in several other surfactants. Suspension in sodium deoxycholate (SDC) or sodium dodecylbenzenesulfonate (SDBS)6,7 generally allowed little or no O2 quenching. However, SWCNT samples in other surfactants, including sodium n-lauroylsarcosine (sarkosyl) and sodium dodecyl sulfate (SDS),8−10 were significantly quenched by dissolved O2. It is possible to limit the coating coverage on nanotube surfaces by using surfactant concentrations too low to support stable SWCNT suspensions. We accordingly prepared samples in 0.05 and 0.1% aqueous SDS and measured their O2 fluorescence quenching spectra. The results, plotted in Figure 2c, clearly show enhanced quenching at the lower surfactant concentration. This supports the interpretation that quenching factors reflect exposure of the nanotube surface. To properly assess the (n,m)-dependent quenching factors for a sample, we first purged it with argon and measured fluorescence spectra induced by three different fixed-wavelength excitation lasers. These measurements were repeated after purging with oxygen. Using the systematic method described previously,11−13 we analyzed the argon-purged spectra to find the set of semiconducting (n,m) species present and their relative abundances. Figure 3 illustrates a typical spectral simulation generated in this way. We then repeated the analysis for the oxygen-purged spectra, keeping all fitting parameters (peak positions, widths, shapes, excitation factors, etc.) the same as for the argon data, except for abundances.

Figure 3. Measured (symbols) and simulated (solid red curve) fluorescence spectra of a bulk SWCNT sample suspended in argonsaturated SDS solution. Thin black curves show the deduced emission components from different semiconducting (n,m) species, the strongest of which are labeled near the top of the frame.

Because these deduced abundance values are proportional to emissive quantum yields, the ratios of apparent abundances with argon to those with oxygen give the desired (n,m)-specific quenching factors. We measured such oxygen quenching factors for SWCNT samples suspended in 10 different ssDNA sequences. Table 1 lists the results, along with those in 0.1% SDS, for six common (n,m) species. Note that quenching values range from 1.08 (indicating minimal effect) to 2.46. Another possible way to estimate coating integrity for SWCNTs in specific ssDNA oligos is by comparing their emission intensities (in air-saturated suspensions) with those of samples coated by a surfactant, such as SDC, thought to provide strong isolation from the environment. As judged by characteristic shifts in the fluorescence peak wavelengths and changes in peak intensity, the addition of SDC solution can completely displace ssDNA from SWCNT surfaces. We have measured the relative intensity increases from SDC displacement for the same combinations of (n,m) species and ssDNA sequences listed in Table 1 (see Figure S3 for an example). Figure 4 shows plots of the correlation between those intensity ratios and the oxygen quenching factors for four different (n,m) species. We find a strong, but not perfect, positive correlation for all species. This suggests that the two effects are closely related and reflect the extent to which the SWCNT surface is isolated from the surrounding solution. 1953

DOI: 10.1021/acs.jpclett.7b00583 J. Phys. Chem. Lett. 2017, 8, 1952−1955

Letter

The Journal of Physical Chemistry Letters

lower quenching factors than the other oligos for (6,5), suggesting specific recognition. Quenching in (TCG)4TC is generally large for species except (8,6). This is consistent with the finding by Tu et al. that (TCG)4TC is a recognition sequence for (8,6) SWCNTs.1 Some interesting patterns can also be seen in Table 1. The (8,6) SWCNT, whose diameter is the largest in the group, shows consistently small quenching factors for all ssDNA sequences. In some cases, quenching factors are nearly a factor of 2 larger for (7,5) than for (8,6), even though these two structures belong to the same (n−m) family and have nearly equal roll-up angles. The difference in their diameters, 0.829 versus 0.966 nm, presumably accounts for most of the large difference in ssDNA surface coverage effectiveness. The effect of ssDNA length can be examined from the last four lines in Table 1, which list quenching factors for (GT) oligos with repeat factors ranging from 10 to 30. Those quenching factors are largest for (GT)10, suggesting that its length may be too short to allow effective wrapping around SWCNTs with diameters smaller than (8,6). The observed factors are smaller and fairly constant for (GT)15 to (GT)30. Finally, we note that our quenching factors are averaged over the two enantiomers of each (n,m) species. Because it is has recently been demonstrated that some ssDNA oligos are sensitive to SWCNT handedness,14 the unresolved enantiomer-specific quenching factors may actually show stronger variations than we observe here. This possibility is currently being investigated, and results will be reported separately.

Table 1. Quenching Factors (Based on Apparent Abundances) for Six Nanotube Species Suspended with 0.1% SDS and 10 ssDNA Sequencesa 0.1% SDS (ATT)4 TTA(TAT)2ATT (TAT)4 (TCG)10 (TCG)4TC (TTA)3TTGTT (GT)30 (GT)20 (GT)15 (GT)10 a

(6,5)

(8,3)

(7,5)

(7,6)

(10,2)

(8,6)

1.13 1.40 1.20 1.39 1.47 1.70 1.47 1.70 1.71 1.67 1.88

1.16 2.17 1.35 1.58 1.85 2.05 2.07 2.03 2.14 1.99 2.33

1.23 1.20 1.44 1.62 1.83 2.02 2.44 1.83 1.93 2.00 2.46

1.15 1.71 1.38 1.45 1.76 1.77 1.64 1.55 1.58 1.56 1.75

1.13 1.47 1.30 1.44 1.59 1.56 1.82 1.51 1.54 1.55 1.73

1.08 1.26 1.16 1.25 1.26 1.10 1.29 1.22 1.17 1.22 1.24

Estimated uncertainties are 0.02.

Our findings in Table 1 can be analyzed to deduce selective wrapping affinities of ssDNA sequences for particular (n,m) species and allow comparison with the results found by Zheng and coworkers in separation experiments. Smaller quenching factors in the Table imply more complete ssDNA wrapping coverage and may point to selective affinity. The oligo sequence (ATT)4 shows a particularly low quenching factor for (7,5), in agreement with Tu et al.’s finding of specific recognition in chromatographic sorting studies. 1 Similarly, the TTA(TAT)2ATT and (TAT)4 sequences (both palindromes) give

Figure 4. Plots for four (n,m) species showing correlations between O2 quenching factors (y axis) and the relative fluorescence intensity enhancements measured when ssDNA is completely displaced by SDC surfactant (x axis). Lines show best-fit linear relations. Red triangles mark data for ssDNA oligos that had previously been reported to have special affinities for the (n,m) species shown in the frame’s upper left-hand corner. 1954

DOI: 10.1021/acs.jpclett.7b00583 J. Phys. Chem. Lett. 2017, 8, 1952−1955

Letter

The Journal of Physical Chemistry Letters

(3) Ao, G.; Khripin, C. Y.; Zheng, M. DNA-Controlled Partition of Carbon Nanotubes in Polymer Aqueous Two-Phase Systems. J. Am. Chem. Soc. 2014, 136 (29), 10383−10392. (4) Cognet, L.; Tsyboulski, D.; Rocha, J.-D. R.; Doyle, C. D.; Tour, J. M.; Weisman, R. B. Stepwise quenching of exciton fluorescence in carbon nanotubes by single-molecule reactions. Science 2007, 316, 1465−1468. (5) Zheng, M.; Diner, B. A. Solution Redox Chemistry of Carbon Nanotubes. J. Am. Chem. Soc. 2004, 126 (47), 15490−15494. (6) Fernandes, R. M. F.; Abreu, B.; Claro, B.; Buzaglo, M.; Regev, O.; Furó, I.; Marques, E. F. Dispersing Carbon Nanotubes with Ionic Surfactants under Controlled Conditions: Comparisons and Insight. Langmuir 2015, 31 (40), 10955−10965. (7) Wenseleers, W.; Vlasov, I. I.; Goovaerts, E.; Obraztsova, E. D.; Lobach, A. S.; Bouwen, A. Efficient Isolation and Solubilization of Pristine Single-Walled Nanotubes in Bile Salt Micelles. Adv. Funct. Mater. 2004, 14 (11), 1105−1112. (8) Richard, C.; Balavoine, F.; Schultz, P.; Ebbesen, T. W.; Mioskowski, C. Supramolecular Self-Assembly of Lipid Derivatives on Carbon Nanotubes. Science 2003, 300 (5620), 775−778. (9) Tummala, N. R.; Striolo, A. SDS Surfactants on Carbon Nanotubes: Aggregate Morphology. ACS Nano 2009, 3 (3), 595−602. (10) Xu, Z.; Yang, X.; Yang, Z. A Molecular Simulation Probing of Structure and Interaction for Supramolecular Sodium Dodecyl Sulfate/ Single-Wall Carbon Nanotube Assemblies. Nano Lett. 2010, 10 (3), 985−991. (11) Weisman, R. B. Fluorimetric characterization of single-walled carbon nanotubes. Anal. Bioanal. Chem. 2010, 396 (3), 1015−1023. (12) Rocha, J.-D. R.; Bachilo, S. M.; Ghosh, S.; Arepalli, S.; Weisman, R. B. Efficient spectrofluorimetric analysis of single-walled carbon nanotube samples. Anal. Chem. 2011, 83, 7431−7437. (13) Bachilo, S. M.; Balzano, L.; Herrera, J. E.; Pompeo, F.; Resasco, D. E.; Weisman, R. B. Narrow (n,m)-distribution of single-walled carbon nanotubes grown using a solid supported catalyst. J. Am. Chem. Soc. 2003, 125 (37), 11186−11187. (14) Ao, G.; Streit, J. K.; Fagan, J. A.; Zheng, M. Differentiating Leftand Right-Handed Carbon Nanotubes by DNA. J. Am. Chem. Soc. 2016, 138 (51), 16677−16685. (15) Cabrerizo, F. M.; Arnbjerg, J.; Denofrio, M. P.; Erra-Balsells, R.; Ogilby, P. R. Fluorescence quenching by oxygen: ″Debunking″ a classic rule. ChemPhysChem 2010, 11, 796−798. (16) Kristiansen, M.; Scurlock, R. D.; Iu, K. K.; Ogilby, P. R. Chargetransfer state and singlet oxygen (1Δg O2) production in photoexcited organic molecule-molecular oxygen complexes. J. Phys. Chem. 1991, 95 (13), 5190−5197.

Further research will be needed to fully understand the mechanism of SWCNT fluorescence quenching by dissolved O2. However, we speculate that a localized singlet exciton generated by light absorption diffuses to a site at which an oxygen molecule has physisorbed onto the nanotube surface.4 Interactions with the triplet (ground state) O2 might then enhance the singlet exciton’s nonradiative recombination in a process analogous to induced internal conversion.15 Another possible exciton decay channel is energy transfer to form 1O2, as occurs when oxygen quenches organic excited states.16 This effect should favor quenching of smaller diameter SWCNTs because of their higher energy excitons, a trend shown in the plot of Figure S4. Either process would reduce the observed fluorescence emission. In summary, we have found that the fluorescence of SWCNTs suspended in water by ssDNA is significantly quenched by dissolved O2. The quenching is quickly and fully reversed at room temperature by displacing the O2, indicating a weak interaction with the nanotubes. The extent of quenching depends on the combination of (n,m) species and ssDNA oligo sequence. This suggests that stronger coating interactions hamper O2 access and reduce quenching. The combinations observed to give low quenching values agree with the specific recognition sequences identified in previous structural sorting studies using ssDNA oligos. We conclude that screening for fluorescence quenching by dissolved oxygen offers a simple and nondestructive way to explore the wrapping interactions of ssDNA with semiconducting SWCNT structures.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpclett.7b00583. pH-dependent quenching spectra, kinetic data, surfactant displacement spectra, and variation of quenching factor with optical band gap. (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: 713-3483709. Fax: 713-3485155. ORCID

R. Bruce Weisman: 0000-0001-8546-9980 Notes

The authors declare the following competing financial interest(s): R.B.W. has a financial interest in Applied NanoFluorescence, LLC, which manufactures instruments used in this study.

■

ACKNOWLEDGMENTS This research was supported by grants from the National Science Foundation (CHE-1409698) and the Welch Foundation (C-0807).

■

REFERENCES

(1) Tu, X.; Manohar, S.; Jagota, A.; Zheng, M. DNA sequence motifs for structure-specific recognition and separation of carbon nanotubes. Nature 2009, 460 (7252), 250−253. (2) Huang, X.; McLean, R. S.; Zheng, M. High-Resolution Length Sorting and Purification of DNA-Wrapped Carbon Nanotubes by SizeExclusion Chromatography. Anal. Chem. 2005, 77 (19), 6225−6228. 1955

DOI: 10.1021/acs.jpclett.7b00583 J. Phys. Chem. Lett. 2017, 8, 1952−1955