We report emission of broadband light in the spectral range 500 nm-900 nm from single walled carbon nanotubes (SWNTs) in a liquid environment upon irradiance by a very low power (typically <5 mW), continuous-wave laser source in a tweezers setup. We show (i) formation of micro-bubbles upon irradiation of fluids containing bundles of SWNTs, (ii) optical trapping of such micro-bubbles, (iii) adhesion of SWNTs on the surface of such micro-bubbles, and (iv) bright emission of white light due to tweezer-induced localized heating of spatially-constrained SWNTs.
©2009 Optical Society of America
Light emission from carbon nanotubes has attracted much attention in recent times  due to potential applications in optoelectronic devices, biomedical agents, and as non-classical sources of light. Various mechanisms of emission have been reported. Current carrying nanotubes are known to undergo resistive heating and emit incandescent light  - a fact used in making light bulbs with filaments or ropes of single and double-walled carbon nanotubes [3,4]. Carbon nanotubes undergoing field emission are known to become luminous, and this has been utilised to devise light sources [5-8]. Electroluminescence, where a small voltage is applied across the ends of a suspended nanotube, is found to give rise to emission of nearinfrared light. Such emission has been attributed either to heating and subsequent thermal emission  or to radiative recombination  of electrons and holes (CNT-LED). Broadband photoluminescence in the infrared and near-infrared has been observed from SWNTs under excitation by lasers operating in the visible and the infrared; these show multiple peaked emission [11-13] arising out of interband electronic transitions. In the case of carbon nanotubes in zeolite templates, irradiation by ultraviolet light has resulted in emission in the near-infrared . In contrast to interband transitions, purely thermal emission is broad and featureless.
To date, all emission has been in the infrared or near-infrared while the excitation sources have generally been of shorter wavelengths. In other words, what has been reported to date is akin to down-conversion. To our knowledge, only one study  has reported laserinduced incandescence in nanotubes yielding light in the visible, but this required large fluences typical of pulsed lasers. Furthermore, most studies on light emission from carbon nanotubes have used isolated or suspended tubes, sometimes thin films, requiring elaborate sample preparation and handling techniques. Aggregated bundles of nanotubes are believed to be poor emitters  because of rapid non-radiative inter-tube relaxation processes within the bundles. The yields increase marginally (by 10%) in individual nanotubes . Current efforts strive to increase the quantum efficiency  even further. In addition, the work reported thus far has invariably involved SWNTs in a dry environment, such as thin films or filaments in vacuum [15,18], while a variety of potential applications demand a fluid environment. Here, we report observations on emission of broadband light in the visible (500 nm–900 nm) from SWNT’s in a liquid environment under irradiance by a very low power continuous wave laser source in a standard optical tweezer. The emission from sample volumes as minute as ~1 µl was visible to the naked eye, and could be well controlled. It is observed that the emission process was preceded by formation of micro-bubbles in the fluid, trapping of such micro-bubbles by tweezer light, and spatial confinement of SWNTs on the surface of such bubbles. Each of these aspects, namely, the formation of micro-bubbles, the confinement of SWNTs to the microbubbles, and white-light generation in the spatially-constrained SWNTs under low power illumination, have opened up the possibility of a variety of applications in diverse situations. For example, controlled formation of microbubbles could be utilized in controlling chemical reactions, aiding metallurgical processes, in bio-medicine and in printing. Micromanipulation of SWNTs and the white light generation from them could be harnessed in directed drug delivery and controlled incandescent illumination, remote light-induced triggers, solar panels, to mention a few.
2. Experimental details
The carbon nanotubes used in our experiments were 1.2-1.5 nm in diameter and were obtained from Sigma Aldrich in the form of bundles. We dispersed these in SDS solution and, following centrifugation, placed them in an optical tweezer set-up that uses a 100X oil-immersed objective with large numerical aperture (NA=1.3) to tightly focus 1064 nm light from a Nd:YVO4 laser to create a trap. Weaker trap depths are achieved with 60X and 20X objectives. In most cases laser power levels as low as 5 mW were used. Light from the irradiated sample was collected via imaging optics, onto a CCD camera interfaced to a computer that enabled real-time recordings to be made. The tweezer setup we used has been described in detail elsewhere ; for the present series of measurements a modification was effected that also permitted simultaneous spectral analysis of light emitted by the sample using a fiber-coupled spectrometer that covers the range 350-1100 nm.
3. Results and discussion
Optical-tweezer-based micro-manipulation of nanotubes has remained elusive as SWNT bundles are strongly repelled from the tweezer focal volume [see Fig. 1(a) (Media 1)]. For particles of size smaller than the wavelength of light, as is the case of the SWNTs being studied, trapping depends on the polarizability of the particle. It is well known that the optical dipole force experienced by a particle in a non-uniform field, E, (as exists in optical tweezers) is given by
The polarizability, α, in the frequency domain, may be expressed as 
In the limit of small damping (γ≪ωo), and away from resonance, the imaginary part is negligible. Then α>0 for ω<ωo, and α<0 for ω>ωo. Thus, red-detuned light attracts the particle to the focus (intensity maximum) while the blue-detuned light repels it.
From the absorption spectrum recorded for our sample (Fig. 1(b)), we find that our tweezer light wavelength (1064 nm) is to the blue of the absorption feature at 1080 nm and thus repels the SWNTs. As indicated by Eq. (2), the absorption peaks in the vicinity of 950 nm, 1040 nm and 1200 nm are of little consequence in this respect as the laser light is considerably more detuned from these.
As a consequence of this repulsion, SWNTs cannot, per se, be optically trapped using conventional tweezers. However, we have discovered a simple method wherein SWNTs can, indeed, be spatially constrained by optical tweezers operating at 1064 nm. This is achieved by means of a bubble-mediated process. While the SWNTs do tend to avoid the focal volume of the trap, it is nevertheless possible to bring them to focal region by moving the translation stage faster than the repelling action experienced by the bundle, and thus subject them to the tweezer light, albeit for short durations.
We have discovered that microbubbles are inevitably formed when SWNTs are exposed to the tweezer light. This bubble formation is a consequence of the large absorption of infrared light by the SWNTs which are then rapidly heated. Dissolved gases and/or fluid vapors within the hot SWNTs undergo expansion and are expelled from the open ends of the SWNTs. Our observation that the bubbles inevitably originate near the end of a carbon nanotube strengthens this surmise. Relatively small bubbles were unstable. Larger bubbles (diameter>0 µm) were stable. It is observed that the micro-bubbles attract proximate SWNTs, which then adhere to the bubble surface, as seen in Fig. 2 (Media 2), presumably due to the non-wetting characteristics of carbon.
The possible role of strong surface tension gradients with accompanying convective currents, and of surface charges in causing the attraction needs to be investigated . The SWNTs which could not earlier be trapped by the tweezer, are now spatially confined to the surface of the microbubble, and are thus rendered micro-manipulable.
When the SWNTs were thus held trapped in the focal volume, sustained emission of light was seen from them, often lasting for as long as the tweezer light was on (Fig. 3(a)). The intensity of this sustained emission was amenable to simple control by varying the incident laser power.
Under tighter focusing, the emission was considerably more intense (Fig. 3(b)), but brief, typically lasting 40-80 ms. Often, in the process of moving the SWNT bundles to the focal volume (prior to bubble formation) momentary flashes of light were seen before the bundles were accelerated away.
Emission from the sample within the tweezer focal volume was intense enough to be visible to the naked eye, both in the case of sustained emission and intense flashes. Visually it tended to be orange-red during sustained emission, and blue/white for intense flashes. We show in Fig. 4 successive frames depicting emission under tight focusing conditions. As the emitting volume becomes bigger and the emitted light becomes brighter, the color changes from red to orange to yellow to violet, and as it dies out, turns to orange again. We attribute these emissions to blackbody-like radiation of the SWNTs that have very efficiently absorbed the tweezer light, and have thus got heated. The progressive change of color with time, and the increase in spot size with heating indicate that this emission is indeed predominantly blackbody-like radiation from local hot-spots. This is also consistent with the observation that the central portion of a well heated spot emits at a shorter wavelength than the periphery, indicating that the flow of energy outward from the hot spot.
A recent report has described the ignition of SWNTs under illumination by a photographic flash, with temperatures of 1500 K being estimated . However, unlike the photographic flash-induced ignition, the incident illumination in our experiment was in the infrared and light emitted by the SWNTs spanned the entire spectrum from 500 nm to 900 nm. Given in Fig. 5 are spectra of intense flashes from SWNT bundles in SDS solution measured at various incident laser powers; the peak intensity of the spectrum measured at 70 mW power was taken to be unity and the remaining spectra were normalized with respect to this. The wavelength scale was calibrated using a standard Hg-Ar source; the intensity scale was normalized to unity at the peak. Attempts to fit a blackbody curve were only partially successful, presumably because of the intervention of additional processes such as self absorption. We note that the peak of the spectrum is at ~570 nm, which would correspond to a temperature of ~5000 K in a blackbody. Our attempts to fit the blackbody equation to our spectrum were unsuccessful. There is more than a single reason for this. Firstly, self-absorption undoubtedly affects the emitted spectrum. Secondly, the relationship between the temperature of our SWNT bundle and the radiation that is emitted is not straightforward and is far from being well understood, as has been cogently discussed in recent work  on light emission from a single carbon nanotube.
We wish to focus here on only the following facet of the spectral information presented in Fig. 5: the data not only illustrate the control we can exercise on emission intensity, but also shows a shift in the peak emission towards the blue with increasing pump power. This constitutes strong evidence in favor of the light being blackbody-like emission, with the temperature of the nanotubes being elevated due to absorption of the tweezer light.
In Fig. 6 we present the spectrum for the same sample under dry conditions, at the same input power as the largest emission in Fig. 5(b). It is seen that under dry conditions, the peak is more towards the blue, indicative of higher temperature; conduction of heat by the surrounding fluid under wet conditions restricts the temperatures attained. While numerous other mechanisms are also known to lead to light emission by nanotubes, we believe that in our case, the emission is predominantly thermal. Earlier observations of emission in SWNTs that utilized pulsed light sources invoked nonlinear processes, which at the tweezer light intensity levels that we have employed, are unlikely.
Radiative recombination, as in a CNT-LED, and interband transitions involving van-Hove singularities may be ruled out, the former due to the fact that no dc bias is applied, and the latter due to the broad featureless nature of the emission. However, we cannot altogether rule out some other possibilities.
For example, it is known that application of electric field can strongly influence the absorption and emission characteristics of carbon nanotubes  via the Stark effect – a distinct possibility at the tweezer focus. Be it as it may, it is the low-power tweezer light that is instrumental in causing the local heating of the nanotubes, the expulsion of the bubbles, the localization of the bubble, and the continued heating of the spatially-constrained nanotubes. Thus our method provides for an efficient conversion of infrared light to visible, by means of directed irradiation onto efficient absorbers that are held immobile by optical means. It may be remarked here that earlier observations of blackbody radiation from SWNTs have been carried out under dry/vacuum conditions, where heat conductive contact was minimal due to the vacuum and the small area of contact between the irregular SWNT bundles and the surrounding (solid) bulk. In the present case, we achieve white light emission in a liquid medium, where significant thermal dissipation would be expected to hinder formation of local hotspots and thus suppress both bubble formation and light emission. Indeed, we have observed that pulsed illumination (using peak powers in excess of 1 GW with 40 fs pulses of 800 nm light) is required for bubble formation when pure liquids (without suspended SWNTs) are used. The energy requirement for the bubble formation is considerably reduced when SWNTs are added – in our case, mW level cw illumination is seen to suffice.
In summary we have demonstrated that weak tweezer light can be made to perform the multiple roles of (i) forming micro-bubbles upon irradiation of fluids containing SWNTs, (ii) trapping and manipulating such micro-bubbles which have SWNTs spatially-constrained on their surface, and (iii) causing localized heating of SWNTs to give rise to white light emission with only mW irradiation.
We are grateful to Jacinta D’Souza for help in sample preparation. We also thank S. Sen and S. Mazumdar for absorption measurements. GR is an Indian Academy of Sciences Summer Fellow from Cochin University of Science and Technology.
References and links
1. P. Avouris, M. Freitag, and V. Perebeinos, “Carbon-nanotube photonics and optoelectronics,” Nat. Photon. 2, 341–350 (2008). [CrossRef]
2. P. Li, K. Jiang, M. Liu, Q. Li, and S. Fan, “Polarized incandescent light emission from carbon nanotubes,” Appl. Phys. Lett. 82, 1763–1765 (2003). [CrossRef]
3. J. Wei, H. Zhu, and D. Wu, “Carbon nanotube filaments in household light bulbs,” Appl. Phys. Lett. 84, 4869–4871 (2004). [CrossRef]
4. C. Li, K. Wang, J. Wei, B. Wei, H. Zhu, Z. Wang, J. Luo, W. Liu, M. Zheng, and D. Wu, “Luminescence of carbon nanotube bulbs,” Chinese Science Bulletin 52, 113–117 (2007). [CrossRef]
5. J. M. Bonard, T. Stöckli, F. Maier, W.A. de Heer, and A. Châtelain, “Field-Emission-Induced Luminescence from carbon nanotubes,” Phys. Rev. Lett. 81, 1441–1444 (1998). [CrossRef]
6. M. Sveningsson, M. Jönsson, O. A. Nerushev, F. Rohmund, and E. E. B. Campbell, “Blackbody radiation from resistively heated multiwalled carbon nanotubes during field emission,” Appl. Phys. Lett. 81, 1095–1097 (2002). [CrossRef]
7. Y. Saito, S. Uemura, and K. Hamaguchi, “Cathode ray tube lighting elements with carbon nanotube field emitters,” Jpn. J. Appl. Phys. 37, L346–348 (1998). [CrossRef]
8. Y. Saito, K. Hata, A. Takakura, J. Yotani, and S. Uemura, “Field emission of carbon nanotubes and its application as electron sources of ultra-high luminance light-source devices,” Physica B 323, 30–37 (2002). [CrossRef]
9. D. Mann, Y. K. Kato, A. Kinkhabwala, E. Pop, J. Cao, X. Wang, L. Zhang, Q. Wang, J. Guo, and H. Dai, “Electrically driven thermal light emission from individual single-walled carbon nanotubes,” Nat. Nanotechnol. 2, 33–38 (2007). [CrossRef]
10. J. A. Misewich, R. Martel, Ph. Avouris, J. C. Tsang, S. Heinze, and J. Tersoff, “Electrically induced optical emission from a carbon nanotube FET,” Science 300, 783–786 (2003). [CrossRef] [PubMed]
11. M. J. O’Connell, S. M. Bachilo, C. B. Huffman, V. C. Moore, M. S. Strano, E. H. Haroz, K. L. Rialon, P. J. Boul, W. H. Noon, C. Kittrell, J. Ma, R. H. Hauge, R. B. Weisman, and R. E. Smalley, “Band gap fluorescence from individual single-walled carbon nanotubes,” Science 297, 593–596 (2002). [CrossRef] [PubMed]
12. S. Lebedkin, F. Hennrich, T. Skipa, and M. Kappes, “Near-infrared photoluminescence of single-walled carbon nanotubes prepared by the laser vaporization method,” J. Phys. Chem. B 107, 1949–1956 (2003). [CrossRef]
13. J. Guo, C. Yang, Z. M. Li, M. Bai, H. J. Liu, G. D. Li, E. G. Wang, C. T. Chan, Z. K. Tang, W. K. Ge, and X. Xiao, “Efficient visible photoluminescence from carbon nanotubes in zeolite templates,” Phys. Rev. Lett. 93, 017402-1-4 (2004) [CrossRef]
14. H. Zeng, C. Yang, J. Dai, and X. Cui, “Light-induced incandescence of SWNT,” J. Phys. Chem. C. 112, 4172–4175 (2008). [CrossRef]
15. R. L. van der Waal, G. M. Berger, T. M. Ticich, and P. D. Patel, “Application of laser-induced incandescence to the detection of carbon nanotubes and carbon nanofibers,” Appl. Opt. 41, 5678–5690 (2002). [CrossRef]
16. F. Wang, G. Dukovic, L. E. Brus, and T. F. Heinz, “Time-resolved fluorescence of carbon nanotubes and its implication for radiative lifetimes,” Phys. Rev. Lett. 92, 177401-1-4 (2004) [CrossRef]
17. S.-Y. Ju, W. P. Kopcha, and F. Papadimitrakopoulos, “Brightly fluorescent single-walled carbon nanotubes via an oxygen-excluding surfactant organization,” Science 323, 1319–1323 (2009). [CrossRef] [PubMed]
18. H. Zeng, C. Yang, J. Dai, and X. Cui, “Light-induced incandescence of single-walled carbon nanotubes,” J. Phys. Chem. C 112, 4172–4175 (2008). [CrossRef]
19. J. A Dharmadhikari and D. Mathur, “Using an optical trap to fold and align single red blood cells,” Curr. Sci. 86, 1432–1437 (2004).
22. P. M. Ajayan, M. Terrones, A. de la Guardia, V. Huc, N. Grobert, B. Q. Wei, H. Lezec, G. Ramanath, and T. W. Ebbesen, “Nanotubes in a flash — ignition and reconstruction,” Science 296, 705 (2002). [CrossRef] [PubMed]
23. Y. Fan, S. B. Singer, R. Bergstrom, and B. C. Regan, “Probing Planck’s law with incandescent light emission from a single carbon nanotube,” Phys. Rev. Lett. 102, 187402-1-4 (2009). [CrossRef]