Open Access
Add to BibSonomyAbstract
Band engineering of transparent semiconductors, such as TiO2, SnO2, has recently been attempted to become black colour in order to increase photocatalytic efficiency in the visible and near infrared ranges. Here, we investigated a previously-neglected property in black oxides by using the example of iron doped indium tin oxide nanoparticles: the photo-thermal effect in the visible and near-infrared ranges. We found a significant enhancement of the photo-thermal generation in the near-infrared biological window that is due to photon absorption, with no significant contribution from surface plasmon resonance. This allows independent magnetic functionalization of the particles by iron doping. These particles may be used for in-vivo cancer treatment where heat is photo-termally generated, while keeping the magnetism as a mean for remote manipulation through permanent magnets.
© 2016 Optical Society of America
1. Introduction
Nanoparticles with efficient absorption in the infrared are used in applications such as photo-thermal therapy (PTT) for cancer treatment [1]. These materials are typical metals, in particular gold. Nevertheless, there remain challenges to circumvent: (1) metal nanoparticles may undergo melting under laser irradiation. (2) their magnetic functionalization for remote transport in biomedical applications is challenging because it requires core-shell structure, with the core being a magnetic metal [2, 3 ]. For this reason, hyperthermia, which makes use of superparamagnetic iron-oxide nanoparticles, is preferred for cancer treatment [4]. In this case, the particles can remotely be driven on the cancer area and thermal energy is generated through hysteresis loss mechanism by applying an alternative field. Yet, superparamagnetism inevitably leads to small hysteresis, and therefore high alternative fields must be applied. (3) photo-thermal effect in iron-oxides is far too weak to be used as an alternative to magnetic heat generation, while keeping the superparamagnetism as a mean for remote manipulation through permanent magnets.
Metal oxide nanoparticles, on the contrary, can easily be functionalized to become magnetic. This is achieved by replacing part of the host cations with 3d magnetic ions. These materials are commonly referred to as “diluted” magnetic oxides because, unlike magnetic metal oxides such as Iron oxides, magnetism is not due to direct exchange interaction. These diluted magnetic oxides belong to the broader family of the diluted magnetic semiconductors (DMS) [5] and can show photo-thermal effect in the near infrared because of surface plasmon resonance if the carrier concentration is high (n > 1020 electrons·cm−3) [6]. For this reason, nanoparticles based on diluted magnetic oxides, such as manganese-doped zinc oxide or iron-tin co-doped indium oxide, have been considered as an alternative to metal nanoparticles [7, 8 ]. Yet, the photo-thermal effect is weak and requires excitation wavelengths longer than those in the first biological window (700 - 980 nm) [1], particularly if the size of the nanoparticles must be large enough to assure a significant magnetic moment for remote manipulation. Diluted magnetic oxides could become a better alternative to core-shell metallic nanoparticles only if photo-thermal generation relied on a physical mechanism other than surface plasmon resonance. If photo-thermal generation does not rely on surface plasmon resonance, the particles can be made larger in size, therefore matching up the magnetic moment of Fe2O3/Au core/shell nanoparticles.
Band engineering of transparent semiconductors, such as TiO2, SnO2 has recently been attempted to increase photocatalytic efficiency in the visible and near infrared ranges [9, 10 ]. A large number of closely energy-spaced defects, usually oxygen vacancies (VO's), can be created in the mid-gap to form a tail band that overlaps with the valence-band. This results in enhanced light absorption, with the material turning from transparent to black, while retaining its chemical stability and the photo-corrosion resistance. Unexpectedly, no significant enhancement of photocatalytic performance has been obtained. While the induced oxygen-vacancies states effectively narrow the band-gap, they also represent efficient recombination centers for photo-generated carriers that will never reach the electrodes. Yet, it has so far been neglected that the strong absorption and fast recombination must turn into a strong photo-thermal effect that can be used, for instance, for photo-thermal therapy.
We have synthesized transparent, as well as black-colour oxide nanoparticles and investigated the enhancement of photo-thermal efficiency in the near-infrared range. VO's are introduced during synthesis by simply increasing the temperature of the process. Valence-band X-ray photoemission spectroscopy and cyclic voltammetry studies revealed the emergence of an energy tail-band in the black nanoparticles that extended from their valence band. The absorption intensity in the near-infrared biological window (700 - 980 nm) was found to be five times larger. In agreement with other authors [11], we found that the large presence of VO's allows easy access to super-paramagnetic functionalization via formation of Fe- VO's F – centres, as revealed by magnetic measurements.
Large quantities of black oxides can readily be obtained by annealing of chemically synthesized transparent nanoparticles [9]. Yet, the post-annealing leads to a non-uniform distribution of VO's along the nanoparticle radius. Besides, freestanding particles cannot be used to understand the physical origin of the optical and photo-thermal properties. Nanoparticle films grown by chemical deposition techniques require annealing in hydrogen to become black, a process that is hazardous and expensive [10]. Therefore, in this research phase of our work we synthesized the nanoparticles by pulsed laser deposition (PLD). This technique allows uniform distribution of VO's over the volume of the nanoparticles by fixing the O2 partial pressure in the vacuum chamber during growth. In addition, PLD offers the possibility to dope the oxide with iron (Fe) for magnetic functionalization. We used black and transparent Fe-doped indium-tin oxide (BFITO and TFITO) nanoparticle films for this study because this technology is readily available in our laboratory and ITO has been proven to be non-toxic [12], but we expect similar results to be obtained with other black oxide semiconductors.
2. Experiment
A single phase dense In1.9-xFexSn0.1O3(x = 0.05) target was prepared via a conventional solid state reaction by mixing and annealing procedure [13,14 ]. The high purity ITO (99.99%; Aldrich) and iron (III) oxide (Fe2O3) (99.995%, Aldrich) powders were mixed with the mentioned atomic ratio and grounded by ball milling for 10 hours. Then the mixed powders were compressed by an oil-compressor at a pressure of 15 Ton into a circular pellet of 2.7 cm in diameter. The pellet was sintered at 1300 °C for 12 hours to form the deposition target. Sapphire was used as the substrate in whole deposition. A laser energy of 300 mJ with a repetition rate of 10 Hz was used to ablate the target in the different vacuum conditions for different kinds of films. The vacuum conditions were kept at high vacuum condition (1 × 10-6 mbar) to prepare BFITO films, while maintained 0.4 mbar with putting O2 gas for TFITO film growths.
Films grown on sapphire substrates in high vacuum conditions were found to be black in colour, whereas introducing 0.4 mbar O2 during deposition was sufficient to obtain a transparent stoichiometric film (see Fig. 1(a) ). The positions of the peaks in the X-ray diffraction patterns (XRD) (Fig. 1(b)) match well with those of the cubic structure of In2O3 (JCPDS card No. 88-2160). The TFITO film shows strong (222), (400) and (444) peaks, similarly to the deposition target. However, in the BFITO film, the (400) direction becomes dominant over the (222). It is known that in texturized, polycrystalline ITO, the change in preferred orientation occurs only between the (400) and the (222) planes [15]. The high ratio between the (400) and (222) peak intensities in the BFITO suggested that the choice of the cubic sapphire substrate employed here favours the c-axis as the preferred growth orientation. This was confirmed by HR-TEM image of nanoparticles separated from the substrate by sonication (Fig. 1(c)). The micrographs of the films (Fig. 1(d)) reveal that increasing the vacuum leads to spontaneously nano-structured films with nanoparticles of diameter ~70 nm and length depending on the growth duration. A similar behaviour has been reported for other semiconductor oxides grown on the same substrate [16]. HAADF-STEM images with the elemental maps (Fig. 1(d)) confirm a highly uniform distribution of the elements, including the magnetic dopant.
Fig. 1 (A) A photograph colour comparison between TFITO and BFITO films; (B) XRD patterns of (A) BFITO film, (B) TFITO film and (c) the Fe-doped ITO target; (C) HR-TEM image of BFITO shows the (400) and (022) crystal lattice planes; (D) The HAADF-STEM image of black Fe-doped ITO nanoparticles. The elemental maps confirm a highly uniform distribution of the elements.
3. Results and discussion
In order to understand the origin of the black color in the nanoparticle films grown under high temperature, we resorted to optical spectroscopy, cyclic voltammetry (CV) and valence-band X-ray photoemission spectroscopy (VB-XPS) for a complete reconstruction of energy bands (Fig. 2 ). The band-gap of the TFITO can readily be obtained from the visible range absorption spectra in Fig. 2(a). We estimated a band gap of 3.1 eV. In order to estimate the band gap of BFITO, we resorted to CV (Fig. 2(b)-2(d)) [17]. We found that the band gap was narrowed to 1.75 eV. VB-XPS was used to measure the densities of state (DOS) in the VB of the two ITO films (Fig. 2(e)). Both films showed VB tails. Yet, for TFITO, the maximum energy is 0.91 eV above the Fermi level, thus the energy between the Fermi level and conduction band (CB) is 2.22 eV. Instead, in BFITO, the tail crosses the Fermi level to be 0.41 eV above it. The reconstructed band diagram is shown in Fig. 2(f).
Fig. 2 (A) Visible range absorbance of (A) BFITO and (b) TFITO; (B-D) Cyclic voltammetry of BFITO; (E) The valence-band XPS spectra of BFITO and TFITO; (F) The schematic illustration of the reconstructed DOS of BFITO and TFITO.
A Fermi level crossing the VB might mistakenly lead to the conclusion that the black oxide is a degenerate p-type doped semiconductor. Hall measurements clearly showed that the film is n-type. The resistivity at T = 300 K was 1.43 × 10−4 Ω·cm, corresponding to a carrier density n = 3 × 1021 cm−3. This means that the VB tail is not a real tail and should be simply regarded as a defect band arising from the large concentration of VO's that overlaps with the VB to narrow the optical band-gap due to Burstein-Moss effect [18].
As always the case for Burstein-Moss shifts of either band edge, the narrowing of the band-gap is only apparent. As a consequence, a much larger optical absorbance in the visible range of BFITO, as compared to TFITO, does not necessarily turn in a similar enhancement of the photocatalytic performance. In fact, we measured the water-oxidation performance of the BFITO and TFITO by using visible light irradiation with a 300 W Xe lamp. In a typical experiment, a 0.2 cm2 film was used as photoanode immersed in 1M KOH electrolyte. As shown in Fig. 3(a) , the TFITO film shows a negligible photo-response upon the overall potential, while the BFITO film shows a stable photo-response. In Fig. 3(b), the photocurrent density is plotted as a function of time at 0.3 V vs. VAg/AgCl. For comparison, the same experiment was carried out under the same condition for a TFITO film (inset Fig. 3(b)). In spite of an increase of current density by about 20% at 0.3 V vs. VAg/AgCl in the case of BFITO (see Fig. 3(a)), the change of on/off photocurrent in Fig. 3(b) is a disappointing 12% due to an offset dark current of 0.25 mA·cm2 . This performance is insufficient to meet solar-driven photocatalysis requirements. Yet, as we demonstrate in the following, the large absorbance in the near-infrared can readily be exploited for photothermal conversion.
Fig. 3 (A) Photocurrent vs. applied potential curves with/without 300W Xe lamp irradiation of BFITO and TFITO. (B) Photocurrent response versus time at 0.3V vs. Ag/AgCl under the visible light irradiation for electrodes fabricated by BFITO. Inset shows the photocurrent response versus time at 0.3V vs. Ag/AgCl under the visible light irradiation for electrodes fabricated by TFITO.
In Fig. 2(a), the absorption of the black film is more than 5 times larger than that of the transparent film in the biological window 700 - 980 nm. It is therefore natural to expect an efficient photo-thermal conversion in the near-infrared range. For this, we used a typical 980 nm laser with power density of 1 W and 1 cm radius spot size to probe the change of temperature under direct irradiation. Figure 4(a) shows the temperature as a function of time over a time-duration of five minutes. The average temperature increase of BFITO is about 65°C, while that of TFITO is only 30°C. For photo-thermal treatment, the typical temperature to destroy tumours is between 50 and 60 °C within the spot size. The images of the temperature distribution in Fig. 4(b) and 4(c) show that the maximum temperature in the case of BFITO can reach 100°C and it remains at around 90°C over an area of the spot size, whereas it barely reaches 50°C at the centre of the spot in TFITO. Considering the inevitable loss of power when travelling through tissues, BFITO is suitable for in-vivo therapy, while TFITO is certainly not.
Fig. 4 (A) Average temperature elevation of BFITO and TFITO as a function of irradiation time under 980 nm laser with the power of 1 W. Temperature distributions of (B) TFITO and (C) BFITO under the same laser irradiation.
In order to understand the potentials of BFITO in bio-applications, photo-thermal conversion efficiency η of BFITO was estimated on free-standing nano-particles. The particles were first separated from the substrate and then preparaed a colloidal solution (50 ppm) according to the protocol described in Ref. 19 [19]. SEM micrographs of the free-standing nano-particle (not shown) show that the particle size is uniform with a mean size of ~100 nm. The colloidal suspension of BFITO nanoparticles with 50 ppm was irradiated by a 980 nm laser with power of I = 0.72 W. The temperature change of the solution under irradiation was recorded for 660 s (Fig. 5 ). The free decay of the temperature was also recorded after switching off the irradiation.
Fig. 5 (A) Absorption spectrum of the aqueous solution of BFITO nanoparticles (50 ppm). (B) The temperature change as a function of time of the same aqueous solution under a laser (980 nm, 0.72 W) irradiation for 660s and temperature decay after switching off the irradiation. (C) Linear time vesus – ln(θ) obtained from the cooling period in Fig. 5(b).
The η of BFITO nanoparticles solution was calculated as [20]:
where h is the heat transfer coefficient, S is the surface area of the container, A980 is the absorbance of the solution at 980 nm, Tmax is the maximum temperature of the colloidal solution after 660s laser irradiation and Tsurr is the environmental temperature. A980 was measured to be 0.0378 (Fig. 5(a)). (Tmax - Tsurr) was measured to be 13.20 °C (Fig. 5(b)). The heat dissipated from the light absorbed by the quartz cell Qdis was measured to be 15.34 mW by using a quartz celling with pure water. The hS product could be calculated by the following equation [21]:where the colloidal solution weight mD = 0.3 g and the heat capacity of solution (we used the heat capacity of water here) CD = 4.2 J/g·°C and the time constant τ s can be obtained by estimating the slope in Fig. 5(c): τ s was calculated to be 517.01 s. Thus, hS = 2.43 mW/ °C.Thus, the photo-thermal conversion efficiency η of BFITO nanoparticle was calculated to be 28.04%, which is relatively high compared with the previous reports, such as ~21.0% of Au nanorods under 800 nm laser [21], 25.7% of Cu9S5 nanocrystals with 980 nm laser [22], and 16.3% for BPDI radical anions under 808 nm laser [23]. The high η of BFITO nanoparticle shows its excellent heat conversion capability.
Let us point out again that photo-thermal effect in our samples is not due to surface plasmon resonance, which occurs at longer wavelengths for particles of our size and our carrier concentration [6]. In addition, doping ITO with Fe reduces the carriers concentration because Fe ions get reduced by acquiring a free electron donated by the Sn, therefore shifting the resonance peak at even longer wavelength [8]. The relaxed constrains on the particle size can be exploited to compensate the loss of magnetic moment in diluted magnetic oxides, as compared to magnetic metals or oxides, by increasing the volume. In our case, we chose to introduce Fe to make the nanoparticles magnetic because a previous work [11] has demonstrated that the magnetic moment in Fe doped semiconductor oxides can approach that of iron oxides if a large number of VO's are present, which is the case in our system in which the black colour is due to a large concentration of VO's This is because VO's are the mediators for exchange interaction in dilute magnetic oxides [11]. For biomedical applications, the nano-particles are required to be superparamagnetic.
Figure 6(a) shows the zero-field-cooling / field-cooling (ZFC/FC) magnetization of our BFITO versus temperature (M vs. T) measured in an external parallel field of H = 50 mT. The M vs. T shows that the particles are superparamagnetic, which is the required behaviour for biomedical applications. There is only one maximum in the ZFC, which indicates that there is only one dominant population of nanoparticles [24], in agreement with the morphological analysis. The maximum of the ZFC, which is a good estimation of the blocking temperature above which the particles are superparamagnetic, is 200 K. Consistently, the magnetization loop at room temperature (Fig. 6(b)) shows little remanence.
Fig. 6 (A) Temperature dependence of the magnetic moment in H = 50 mT after zero field cooling (ZFC) and field cooling (FC). (B) Magnetization loop at room temperature.
4. Conclusion
In conclusion, we have demonstrated that black iron doped indium tin oxide nanoparticles show high photo-thermal effect in the near-infrared range, which is due to photon absorption and not surface plasmon resonance. In addition, they can be easily functionalized to become superparamagnetic, which may provide a valid alternative to magnetic metal / Au core / shell nanoparticles for future biology applications, such as cancer treatment.
Acknowledgments
This study was supported by the Research Grants Council of the Hong Kong Special Administrative Region, China (CityU 104512), and by the National Natural Science Foundation of China (Grant No. 11274261). Financial support from CityU [7004388] is also acknowledged.
References and links
1. D. Jaque, L. Martínez Maestro, B. del Rosal, P. Haro-Gonzalez, A. Benayas, J. L. Plaza, E. Martín Rodríguez, and J. García Solé, “Nanoparticles for photothermal therapies,” Nanoscale 6(16), 9494–9530 (2014). [CrossRef] [PubMed]
2. W.-R. Lee, M. G. Kim, J.-R. Choi, J.-I. Park, S. J. Ko, S. J. Oh, and J. Cheon, “Redox-transmetalation process as a generalized synthetic strategy for core-shell magnetic nanoparticles,” J. Am. Chem. Soc. 127(46), 16090–16097 (2005). [CrossRef] [PubMed]
3. C. S. Levin, C. Hofmann, T. A. Ali, A. T. Kelly, E. Morosan, P. Nordlander, K. H. Whitmire, and N. J. Halas, “Magnetic-plasmonic core-shell nanoparticles,” ACS Nano 3(6), 1379–1388 (2009). [CrossRef] [PubMed]
4. F. Canfarotta and S. A. Piletsky, “Engineered magnetic nanoparticles for biomedical applications,” Adv. Healthc. Mater. 3(2), 160–175 (2014). [CrossRef] [PubMed]
5. T. Dietl, “A ten-year perspective on dilute magnetic semiconductors and oxides,” Nat. Mater. 9(12), 965–974 (2010). [CrossRef] [PubMed]
6. M. Kanehara, H. Koike, T. Yoshinaga, and T. Teranishi, “Indium tin oxide nanoparticles with compositionally tunable surface plasmon resonance frequencies in the near-IR region,” J. Am. Chem. Soc. 131(49), 17736–17737 (2009). [CrossRef] [PubMed]
7. N. S. Norberg, K. R. Kittilstved, J. E. Amonette, R. K. Kukkadapu, D. A. Schwartz, and D. R. Gamelin, “Synthesis of colloidal Mn2+:ZnO quantum dots and high-TC ferromagnetic nanocrystalline thin films,” J. Am. Chem. Soc. 126(30), 9387–9398 (2004). [CrossRef] [PubMed]
8. G. S. Shanker, B. Tandon, T. Shibata, S. Chattopadhyay, and A. Nag, “Doping controls plasmonics, electrical conductivity, and carrier-mediated magnetic coupling in Fe and Sn codoped In2O3 nanocrystals: local structure is the key,” Chem. Mater. 27(3), 892–900 (2015). [CrossRef]
9. X. Chen, L. Liu, P. Y. Yu, and S. S. Mao, “Increasing solar absorption for photocatalysis with black hydrogenated titanium dioxide nanocrystals,” Science 331(6018), 746–750 (2011). [CrossRef] [PubMed]
10. G. Wang, H. Wang, Y. Ling, Y. Tang, X. Yang, R. C. Fitzmorris, C. Wang, J. Z. Zhang, and Y. Li, “Hydrogen-treated TiO2 nanowire arrays for photoelectrochemical water splitting,” Nano Lett. 11(7), 3026–3033 (2011). [CrossRef] [PubMed]
11. J. M. D. Coey, A. P. Douvalis, C. B. Fitzgerald, and M. Venkatesan, “Ferromagnetism in Fe-doped SnO2 thin films,” Appl. Phys. Lett. 84(8), 1332 (2004). [CrossRef]
12. R. Liu, H. Y. Zhang, Z. X. Ji, R. Rallo, T. Xia, C. H. Chang, A. Nel, and Y. Cohen, “Development of structure-activity relationship for metal oxide nanoparticles,” Nanoscale 5(12), 5644–5653 (2013). [CrossRef] [PubMed]
13. X. L. Wang, K. H. Lai, and A. Ruotolo, “A comparative study on the ferromagnetic properties of undoped and Mn-doped ZnO,” J. Alloys Compd. 542, 147–150 (2012). [CrossRef]
14. X. L. Wang, C. Y. Luan, Q. Shao, A. Pruna, C. W. Leung, R. Lortz, J. A. Zapien, and A. Ruotolo, “Effect of the magnetic order on the room-temperature band-gap of Mn-doped ZnO thin films,” Appl. Phys. Lett. 102(10), 102112 (2013). [CrossRef]
15. P. Thilakan and J. Kumar, “Studies on the preferred orientation changes and its influenced properties on ITO thin films,” Vacuum 48(5), 463–466 (1997). [CrossRef]
16. Q. Shao, P. S. Ku, X. L. Wang, W. F. Cheng, J. A. Zapien, C. W. Leung, F. Borgatti, A. Gambardella, V. Dediu, R. Ciprian, and A. Ruotolo, “Chemical states and ferromagnetism in heavily Mn-substituted zinc oxide thin films,” J. Appl. Phys. 115(15), 153902 (2014). [CrossRef]
17. S. N. Inamdar, P. P. Ingole, and S. K. Haram, “Determination of band structure parameters and the quasi-particle gap of CdSe quantum dots by cyclic voltammetry,” ChemPhysChem 9(17), 2574–2579 (2008). [CrossRef] [PubMed]
18. M. Munoz, F. H. Pollak, M. Kahn, D. Ritter, L. Kronik, and G. M. Cohen, “Burstein-Moss shift of n-doped In0.53Ga0.47As/InP,” Phys. Rev. B 63(23), 233302 (2001). [CrossRef]
19. K. E. Tettey, J. W. C. Ho, and D. Lee, “Modulating layer-by-layer assembly of oppositely charged nanoparticles Using a short amphiphilic molecule,” J. Phys. Chem. C 115(14), 6297–6304 (2011). [CrossRef]
20. C. M. Hessel, V. P. Pattani, M. Rasch, M. G. Panthani, B. Koo, J. W. Tunnell, and B. A. Korgel, “Copper selenide nanocrystals for photothermal therapy,” Nano Lett. 11(6), 2560–2566 (2011). [CrossRef] [PubMed]
21. X. Liu, B. Li, F. Fu, K. Xu, R. Zou, Q. Wang, B. Zhang, Z. Chen, and J. Hu, “Facile synthesis of biocompatible cysteine-coated CuS nanoparticles with high photothermal conversion efficiency for cancer therapy,” Dalton Trans. 43(30), 11709–11715 (2014). [CrossRef] [PubMed]
22. Q. Tian, F. Jiang, R. Zou, Q. Liu, Z. Chen, M. Zhu, S. Yang, J. Wang, J. Wang, and J. Hu, “Hydrophilic Cu9S5 Nanocrystals: A Photothermal Agent with a 25.7% Heat Conversion Efficiency for Photothermal Ablation of Cancer Cells in Vivo,” ACS Nano 5(12), 9761–9771 (2011). [CrossRef] [PubMed]
23. Y. Jiao, K. Liu, G. T. Wang, Y. P. Wang, and X. Zhang, “Supramolecular free radicals: near-infrared organic materials with enhanced photothermal conversion,” Chem. Sci. (Camb.) 6(7), 3975–3980 (2015). [CrossRef]
24. D. Peddis, C. Cannas, A. Musinu, and G. Piccaluga, “Coexistence of superparamagnetism and spin-glass like magnetic ordering phenomena in a CoFe2O4-SiO2 Nanocomposite,” J. Phys. Chem. C 112(13), 5141–5147 (2008). [CrossRef]
