- Email: [email protected]
Effect of liquid level and laser power on the formation of spherical alumina nanoparticles by nanosecond laser ablation of alumina target
Thin Solid Films 523 (2012) 46–51
Contents lists available at SciVerse ScienceDirect
Thin Solid Films journal homepage: www.elsevier.com/locate/tsf
Effect of liquid level and laser power on the formation of spherical alumina nanoparticles by nanosecond laser ablation of alumina target Sharif Abdullah Al-Mamun a, Reiko Nakajima b, Takamasa Ishigaki a, b, c, d,⁎ a
Research Center for Micro-Nano Technology, Hosei University, 3-11-15 Midori-cho, Koganei, Tokyo 184‐0003, Japan Department of Materials Chemistry, Hosei University, 3-7-2 Kajino-cho, Koganei, Tokyo 184‐8584, Japan Department of Chemical Science and Technology, 3-7-2 Kajino-cho, Koganei, Tokyo 184‐8584, Japan d Nano Ceramics Center, National Institute for Materials Science, 1‐1 Namiki, Tsukuba, Ibaraki 305‐0044, Japan b c
a r t i c l e
i n f o
Available online 12 June 2012 Keywords: Nanosecond laser ablation Alumina Liquid plasma Size distributions
a b s t r a c t Alumina nanoparticles (NPs) were synthesized by laser ablation of a bulk α-alumina (corundum) target immersed in distilled water using nanosecond laser pulses of 1064-nm wavelength. We investigated the effect of laser power and water column above the target. Synthesized particles were analyzed regarding particle shape and size distributions with scanning electron and transmission electron microscopy. Ablated NPs were spherical in shape and the average particle size ranged from 12 to 18 nm at varied laser power and water levels, although a very small number of melted droplets of submicron spheroids and irregularshaped cracked particles were observed. X-ray diffraction analysis was conducted, which shows mainly the peaks of α-Al2O3 and minor peaks of γ-Al2O3. Phase identification of NPs, using high-resolution transmission electron micrograph lattice images and fast Fourier transform exhibits both metastable γ-Al2O3 and stable α-Al2O3 phases. © 2012 Elsevier B.V. All rights reserved.
1. Introduction Aluminum oxide nanoparticles (NPs) are prepared for a variety of applications due to their large surface area, which gives improved characteristics including catalytic activity compared with bulk alumina. Recent studies have shown that nanofluids (aluminum oxide NPs in water) have significantly greater transport properties of thermal and electrical conductivity and viscosity compared to their base fluids due to large surface area to volume ratio and certain effects of Brownian motion of the NPs [1]. Unlike bulk materials, the principal parameters of NPs are their shape, size, and morphological substructure of the substance, which primarily define the physical and chemical properties. Laser ablation in liquid has been one of the most popular methods for NP synthesis. Although Al2O3 NPs can be generated by a variety of techniques, such as the chemical method [2,3], plasma or flame synthesis [4,5], ball milling [6], electric spark discharge and sono-hydrolysis [7], contamination by constituting reactants is often inevitable. It has recently been reported that laser ablation in liquid initiates a higher ablation rate than that in a gas phase [8], which can be better explained by confined plasma with higher pressure [9] than that in ambient air. Because the induced shock wave lasts three times longer in liquid than that in air, mechanical responses of substrates tend to be enhanced. ⁎ Corresponding author at: Department of Chemical Science and Technology, 3-7-2 Kajino-cho, Koganei, Tokyo 184‐8584, Japan. Tel./fax: + 81 42 387 6134. E-mail address: [email protected] (T. Ishigaki). 0040-6090/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.tsf.2012.06.011
The liquid level and laser power could also be optimized to control the ablation rate and particle size distributions [10]. The synthesis of alumina NPs by using the liquid phase laser ablation method has been reported. Khan et al. [11] synthesized γ-Al2O3 using continuous wave laser ablation in liquid with the average size ranging from 17 to 29 nm. Sajti et al. [12,13] achieved high NP productivity of 1.3 g/h at 18.5 W of focused laser power at a 4-kHz repetition rate by aiding the liquid flow and scanning the target, but the average size was comparatively large, around 30 nm. Liu et al. [14] synthesized γAl2O3 and its derivative θ-type structure by irradiating an Al metal target in the liquid phase. Musaev et al. [15] irradiated a bulk corundum (α-Al2O3) target with UV radiation (337 nm) but unfortunately, most of the ablated particles were submicron and micron-sized. In this work, spherical Al2O3 nanocondensates (12–18 nm) were synthesized by irradiating a bulk corundum target with an infrared pulsed laser (1064 nm) immersed in water. We varied the ablation rate and size distributions by controlling laser power and liquid layer thickness in nanosecond laser ablation of the alumina target. We also investigated the crystal phase of the ablated NPs. 2. Experimental details Ablation experiments were carried out with a Q-switched Nd:YAG (Quantel Brilliant b; Les Ulis, France) pulsed laser source, which provides 6-ns-pulse 1064-nm wavelength with a repetition rate of 10 Hz. Fig. 1 is a schematic of the experimental setup with a laser source
S.A. Al-Mamun et al. / Thin Solid Films 523 (2012) 46–51
Fig. 1. Schematic of the experimental setup.
along with laser guiding and focusing mirrors and lens system. The laser beam, which was in the TEM00 mode, was focused onto a laser spot of approximately 250 μm. The size of the laser spot was
47
measured by irradiating a specially made burn paper placed along the target surface. The irradiated corundum target was a 10 × 10 × 4 mm sintered body with 99.9% purity and a relative density above 99%. All ablation experiments were conducted at room temperature in atmospheric pressure in a distilled water environment. The target crystal was immersed in water contained in a small 20-ml beaker which was mounted on a motor-driven rotator (0–100 rpm). The purpose of rotation was to ensure uniform irradiation on the target and the movement of water to enhance the diffusion of the ablated particles. The rotation speed was set at 40 rpm. The laser beam entered the solution from above at a normal incident angle to the target. The thickness of the distilled water layer above the target (Fig. 1) was varied to investigate the effect on the ablation rate and particle size. The beam irradiated the target for 1–2 h. Laser power was also varied and the effect of laser fluence on the ablation rate, size, and morphology of the synthesized NPs was observed. Laser power was measured with a low-power thermal sensor (Ophir; Model 7Z01560; Tokyo, Japan). Laser-generated particle size, and the distribution and morphology of water-dispersed ceramic NPs were determined using scanning electron microscopy (SEM; JEOL JSM-5310, 15 kV; Tokyo, Japan) and transmission electron microscopy (TEM; JEOL JEM2100F, 200 kV; Tokyo, Japan). Crystal phase was also analyzed with an X-ray diffractometer
Fig. 2. (a) SEM micrograph at a water level of 2 mm, (b) and (c) TEM images at a water level of 4 and 6 mm, respectively; (d) histogram showing NP size distributions at 6 mm; laser power was set to 80 mW [fluence; 20 J/cm2].
48
S.A. Al-Mamun et al. / Thin Solid Films 523 (2012) 46–51
and laser fluence of 3.1 J/cm 2. In our experiment, the ablation rate was comparatively very high at a water level of only 2 mm, and it was found that irregular-shaped submicron- and micron-sized crater particles tend to dominate causing a high ablation weight, as shown in Fig. 2(a). This maximum ablation rate at a minimum water level can be attributed to fragmentation associated with melting of the target when the fluence is 20 J/cm 2. Fragmentation occurred due to lattice heat propagation, lattice expansion, thermally induced stress and brittle fracture [16], and shock-wave-induced mechanical vibration. At a water level of 4 mm, the ablation rate sharply decreased due to the absence of large irregular-shaped particles, indicating that mechanical stress was no longer present and all ablated particles were nano-sized and spherical resulting from melting, evaporation, and condensation. At higher water levels, the ablation rate gradually decreased; probably due to an increase in loss of incident power absorbed by the water and the already ablated particles present on the laser path. Thus, the optimum value of water layer thickness divides the water-level domain into two zones. Both fragmentation and melting occur at a water level below the optimum value, whereas only melting occurs above that value. Particle size (both d50 and average) tends to decrease with increasing water level, as shown in Fig. 3(b). The laser light scattering experiment conducted by Soliman et al. [17,18], and Sasaki and Takada [19] indicated the growth of particles inside the cavitation bubble, which actually contains the plume. They also found that the dynamics of the cavitation bubble, thus the size of the particles, could be controlled by water pressure. There is also a complex interaction among collision frequency of particles inside the plume; coalescence and shock wave intensity may define particle size at different water levels. The decrease in size at a higher water level could also be attributed to laser absorption by water, which could affect the plume temperature and collision frequency of particles. 3.2. Effect of laser power Fig. 3. (a) Dependence of ablation rate and (b) diameter of the synthesized NPs on a water level above the alumina target.
with CuKα1 radiation (RIGAKU Smartlab, 40 kV, 30 mA, wavelength 0.1544 nm; Tokyo, Japan). 3. Results and discussion Laser fluence, liquid level above the target, and laser spot size play an important role in determining the shape, size, morphology, phase, and ablation efficiency of the synthesized particles. 3.1. Effect of liquid level above target To investigate the effect of the liquid level on the particle size and productivity, the water level above the target was varied from 2 to 10 mm while keeping other parameters constant (power was kept at 80 mW, fluence; 20 J/cm2). Fig. 2(a) is an SEM micrograph of the ablated particles at a water level of 2 mm. Panels (b) and (c) of Fig. 2 are the TEM images of ablated NPs at a water level of 4 and 6 mm, respectively, while Fig. 2(d) shows the particle size distributions at 6 mm. The calculated mass-median-diameter (d50) was 28.5 nm and the arithmetic average of the NPs was 12.3 nm. ImageJ software was used and particle sizes ranging from 1 to 60 nm, which were clearly identified from the TEM micrographs, were taken into account when measuring NP size. Fig. 3(a) shows the ablation weight per hour at different water levels. According to Zhu et al. [9], ablation in a water-confined environment is greatly affected by the induced shock wave triggered by adiabatically expanding plasma; thus, the ablation rate should be maximum at the optimum water level. They found that the ablation rate was maximum at a water level of 1.1 mm by irradiating a Si target with a 248-nm KrF excimer laser with a pulse duration of 23 ns
Alumina, a wide band gap material, has a very narrow range of active laser fluence for synthesizing NPs. Laser power above the corresponding fluence causes mechanical stress; thus, large irregular-shaped cracked particles, while below that range, results in insufficient energy for ablation. For this reason, the choice of suitable fluence is an important factor. Panels (a) and (b) of Fig. 4 are the TEM images at laser powers of 80 and 120 mW. With a laser spot size of 250 μm, the calculated laser fluences at 80 and 120 mW were 20 and 30 J/cm2, respectively. In all the cases, the water level was kept to 4 mm. Fig. 4(c) is an SEM micrograph of the ablated particles at a laser power of 200 mW (fluence; 50 J/cm2). At laser fluences of 20 and 30 J/cm2, synthesized nanocondensates were less than 100 nm but at a fluence of 50 J/cm2, large spherical droplets of submicron- and even micron-sized particles were observed, which could be attributed to extreme coalescence and agglomeration of particles. Note that only spherical particles were observed even at high power, indicating that melting only occurs at this water level. Fig. 4(d) shows the variation in the ablation rate as a function of laser power at a water level of 4 mm. As expected, the ablation rate increased with power; thus, the laser fluence and particle size increased as suggested from the histograms in Fig. 5(a) and (b) for 80 and 120 mW, respectively. Fig. 5(c) also suggests that the particle diameter increases with incident laser power. At high power, temperature and pressure inside the plasma plume increase; this further intensifies the collision frequency among particles, leading to the coalescence and formation of large particles. 3.3. Crystal phase identification Fig. 6 shows X-ray diffraction patterns of (a) bulk α-Al2O3, and (b) Al2O3 NPs synthesized by laser ablation in water at a laser power of 80 mW (20 J/cm 2) and a water level of 4 mm, indicating
S.A. Al-Mamun et al. / Thin Solid Films 523 (2012) 46–51
49
Fig. 4. Electron micrographs (a), (b), and (c) at laser powers of 80, 120, and 200 mW, respectively, and at a water level of 4 mm, (d) dependence of ablation rate on laser power.
the mainly stable α-Al2O3 phase with one γ-Al2O3 peak at the (220) plane. It is evident that some laser generated NPs have identical trigonal Bravais lattice structures as the solid target, corresponding to the α-phase alumina with space group R3c, while others evaporated and nucleated particles agglomerated as the thermodynamically metastable γ-phase after solidification. Phase identification for individual particles was also done using high-resolution TEM (HRTEM) and fast Fourier transform (FFT) images. Small synthesized particles were spherical and single crystals with α or γ-phase, but relatively large particles were a result of coalescence of two or more small particles of the same phase showing multiple crystal planes. For example, panels (a) and (b) of Fig. 7 are the HRTEM, and FFT of the corresponding HRTEM images of a spherical particle of about 30 nm. The interplaner spacing (d-spacing) of various planes was calculated and compared with JCPDS, PDF2 database (release 2010), and the γ-phase was identified. HRTEM revealed a d-spacing of 0.46 and 0.24 nm for the corresponding crystal planes of (111) and (311), respectively, for γ-Al2O3. The FFT of the HRTEM image shows γ(111), γ(311), and γ(400) crystal planes. Panels (c) and (d) of Fig. 7 are the HRTEM and FFT images of another particle of almost the same size, which confirms α-Al2O3. The growth of the α(113) and α(104) planes with a tilting angle of 39.6° was observed on the FFT image. Determination of the crystal phase is a complex function of the particle cooling rate (especially in the liquid phase), specific surface area of NPs, and free energy of particles. From a thermodynamic point of view, the α-phase should be stable for the coarsely crystalline Al2O3. However, the γ-phase becomes more stable at the nanoscale due to its lower
surface energy; nevertheless, it exhibits higher entropy than α-Al2O3 [2]. McHale et al. [2] predicted that γ-Al2O3 should become an energetically stable crystal structure for a specific particle surface area exceeding 125 m2 g− 1 at room temperature. Sajti et al. [13] calculated the specific surface area with the following equation: ATot ¼ VMpAbl ⋅ρ Ap , where MAbl is the total ablated mass, ρis the density of the corundum, and Vp and Ap are the volume and effective surface area of a single spherical particle. They found stable α-phase for a particle size of 30 nm. According to this assumption, alumina NPs larger than 12 nm, with a density of 3.97 g/cm 3, should be of a stable α-phase. However, there has not been any strict boundary of the crystal phase of particles defined in terms of size. For example, Khan et al. [11] found their synthesized NPs to be γ-phase, even at a particle size of 29 nm. In our work, we found both α and γ-phases to be present at an average particle size between 12 and 18 nm. We did not find a very straight relationship between particle size and phase since both the stable α-phase and metastable γ-phase were observed for the same particle size. 4. Conclusion We investigated the dependence of synthesized NP productivity, size, and morphology on laser power and liquid level. The synthesized particles were spherical with an average size between 12 and 18 nm (d50; 26–34 nm) depending on the laser operating parameters. The formation mechanism involved evaporation, melting, and dissociation followed by chemical reactions, recombination and crystallization of NPs.
50
S.A. Al-Mamun et al. / Thin Solid Films 523 (2012) 46–51
(a)
(b)
(c)
Fig. 6. Crystal structure: X-ray diffraction patterns of (a) bulk α-Al2O3 target, and (b) Al2O3 NPs synthesized by laser ablation in water for a laser power of 80 mW and a water level of 4 mm, indicating a stable α-Al2O3 phase.
35 References
Diameter [nm]
30 d 50 25 20 Arithmetic average 15 10 80
90
100
110
120
Laser power [mW] Fig. 5. Histograms (a), and (b) showing NP size distributions at 80 and 120 mW, respectively, and (c) dependence of NP size on the incident laser power at a water level of 4 mm.
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19]
K. Fui, V. Wong, T. Kurma, Nanotechnology 19 (2008) 345702. J.M. McHale, A. Auroux, J. Perrotta, A. Navrotsky, Science 277 (1997) 788. D.J. Suh, T.-J. Park, J.-H. Kim, K.-L. Kim, Chem. Mater. 9 (1997) 1903. J.R. Jensen, T. Johannessen, S. Wedel, H. Livbjerg, J. Nanopart. Res. 2 (2000) 363. T. Johannessen, S.E. Pratsinis, H. Livbjerg, Chem. Eng. Sci. 55 (2000) 177. Y.Q. Wu, Y.F. Zhang, X.X. Huang, J.K. Guo, Ceram. Int. 27 (2001) 265. R. Sivarajan, S. Elena, C. Baruch, C. Rachman, G. Aharon, J. Am. Ceram. Soc. 83 (2000) 89. N. Bärsch, A. Gatti, S. Barcikowski, J. Laser Micro/Nanoeng. 4 (2007) 66. S. Zhu, Y.F. Lu, M.H. Hong, X.Y. Chen, J. Appl. Phys. 89 (2001) 2004. S. Zhu, Y.F. Lu, M.H. Hong, Appl. Phys. Lett. 79 (2001) 1396. S.Z. Khan, Z. Liu, L. Li, Appl. Phys. A 101 (2010) 781. C.L. Sajti, R. Sattari, B. Chichkov, S. Barcikowski, Appl. Phys. A 100 (2010) 203. C.L. Sajti, R. Sattari, B.N. Chikov, S. Barcikowski, J. Phys. Chem. C 114 (2010) 2421. I.L. Liu, P. Shen, S.Y. Chen, J. Phys. Chem. C 114 (2010) 7751. O.R. Musaev, A.E. Midgley, J.M. Wrobel, M.B. Kruger, Chem. Phys. Lett. 487 (2010) 81. R.L. Webb, L.C. Jensen, S.C. Langford, J.T. Dickinson, J. Appl. Phys. 74 (1993) 2338. W. Soliman, N. Takada, K. Sasaki, Appl. Phys. Express 3 (2010) 035201. W. Soliman, N. Takada, K. Sasaki, Jpn. J. Appl. Phys. 50 (2011) 108003. K. Sasaki, N. Takada, Pure Appl. Chem. 82 (2010) 1317.
S.A. Al-Mamun et al. / Thin Solid Films 523 (2012) 46–51
51
Fig. 7. HRTEM images (a), and (c) showing a crystallized particle; magnification of the marked area (inset) shows interplaner spacing, and (b), (d) are the corresponding FFT images showing spots matching γ-Al2O3 and α-Al2O3, respectively.
Contents lists available at SciVerse ScienceDirect
Thin Solid Films journal homepage: www.elsevier.com/locate/tsf
Effect of liquid level and laser power on the formation of spherical alumina nanoparticles by nanosecond laser ablation of alumina target Sharif Abdullah Al-Mamun a, Reiko Nakajima b, Takamasa Ishigaki a, b, c, d,⁎ a
Research Center for Micro-Nano Technology, Hosei University, 3-11-15 Midori-cho, Koganei, Tokyo 184‐0003, Japan Department of Materials Chemistry, Hosei University, 3-7-2 Kajino-cho, Koganei, Tokyo 184‐8584, Japan Department of Chemical Science and Technology, 3-7-2 Kajino-cho, Koganei, Tokyo 184‐8584, Japan d Nano Ceramics Center, National Institute for Materials Science, 1‐1 Namiki, Tsukuba, Ibaraki 305‐0044, Japan b c
a r t i c l e
i n f o
Available online 12 June 2012 Keywords: Nanosecond laser ablation Alumina Liquid plasma Size distributions
a b s t r a c t Alumina nanoparticles (NPs) were synthesized by laser ablation of a bulk α-alumina (corundum) target immersed in distilled water using nanosecond laser pulses of 1064-nm wavelength. We investigated the effect of laser power and water column above the target. Synthesized particles were analyzed regarding particle shape and size distributions with scanning electron and transmission electron microscopy. Ablated NPs were spherical in shape and the average particle size ranged from 12 to 18 nm at varied laser power and water levels, although a very small number of melted droplets of submicron spheroids and irregularshaped cracked particles were observed. X-ray diffraction analysis was conducted, which shows mainly the peaks of α-Al2O3 and minor peaks of γ-Al2O3. Phase identification of NPs, using high-resolution transmission electron micrograph lattice images and fast Fourier transform exhibits both metastable γ-Al2O3 and stable α-Al2O3 phases. © 2012 Elsevier B.V. All rights reserved.
1. Introduction Aluminum oxide nanoparticles (NPs) are prepared for a variety of applications due to their large surface area, which gives improved characteristics including catalytic activity compared with bulk alumina. Recent studies have shown that nanofluids (aluminum oxide NPs in water) have significantly greater transport properties of thermal and electrical conductivity and viscosity compared to their base fluids due to large surface area to volume ratio and certain effects of Brownian motion of the NPs [1]. Unlike bulk materials, the principal parameters of NPs are their shape, size, and morphological substructure of the substance, which primarily define the physical and chemical properties. Laser ablation in liquid has been one of the most popular methods for NP synthesis. Although Al2O3 NPs can be generated by a variety of techniques, such as the chemical method [2,3], plasma or flame synthesis [4,5], ball milling [6], electric spark discharge and sono-hydrolysis [7], contamination by constituting reactants is often inevitable. It has recently been reported that laser ablation in liquid initiates a higher ablation rate than that in a gas phase [8], which can be better explained by confined plasma with higher pressure [9] than that in ambient air. Because the induced shock wave lasts three times longer in liquid than that in air, mechanical responses of substrates tend to be enhanced. ⁎ Corresponding author at: Department of Chemical Science and Technology, 3-7-2 Kajino-cho, Koganei, Tokyo 184‐8584, Japan. Tel./fax: + 81 42 387 6134. E-mail address: [email protected] (T. Ishigaki). 0040-6090/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.tsf.2012.06.011
The liquid level and laser power could also be optimized to control the ablation rate and particle size distributions [10]. The synthesis of alumina NPs by using the liquid phase laser ablation method has been reported. Khan et al. [11] synthesized γ-Al2O3 using continuous wave laser ablation in liquid with the average size ranging from 17 to 29 nm. Sajti et al. [12,13] achieved high NP productivity of 1.3 g/h at 18.5 W of focused laser power at a 4-kHz repetition rate by aiding the liquid flow and scanning the target, but the average size was comparatively large, around 30 nm. Liu et al. [14] synthesized γAl2O3 and its derivative θ-type structure by irradiating an Al metal target in the liquid phase. Musaev et al. [15] irradiated a bulk corundum (α-Al2O3) target with UV radiation (337 nm) but unfortunately, most of the ablated particles were submicron and micron-sized. In this work, spherical Al2O3 nanocondensates (12–18 nm) were synthesized by irradiating a bulk corundum target with an infrared pulsed laser (1064 nm) immersed in water. We varied the ablation rate and size distributions by controlling laser power and liquid layer thickness in nanosecond laser ablation of the alumina target. We also investigated the crystal phase of the ablated NPs. 2. Experimental details Ablation experiments were carried out with a Q-switched Nd:YAG (Quantel Brilliant b; Les Ulis, France) pulsed laser source, which provides 6-ns-pulse 1064-nm wavelength with a repetition rate of 10 Hz. Fig. 1 is a schematic of the experimental setup with a laser source
S.A. Al-Mamun et al. / Thin Solid Films 523 (2012) 46–51
Fig. 1. Schematic of the experimental setup.
along with laser guiding and focusing mirrors and lens system. The laser beam, which was in the TEM00 mode, was focused onto a laser spot of approximately 250 μm. The size of the laser spot was
47
measured by irradiating a specially made burn paper placed along the target surface. The irradiated corundum target was a 10 × 10 × 4 mm sintered body with 99.9% purity and a relative density above 99%. All ablation experiments were conducted at room temperature in atmospheric pressure in a distilled water environment. The target crystal was immersed in water contained in a small 20-ml beaker which was mounted on a motor-driven rotator (0–100 rpm). The purpose of rotation was to ensure uniform irradiation on the target and the movement of water to enhance the diffusion of the ablated particles. The rotation speed was set at 40 rpm. The laser beam entered the solution from above at a normal incident angle to the target. The thickness of the distilled water layer above the target (Fig. 1) was varied to investigate the effect on the ablation rate and particle size. The beam irradiated the target for 1–2 h. Laser power was also varied and the effect of laser fluence on the ablation rate, size, and morphology of the synthesized NPs was observed. Laser power was measured with a low-power thermal sensor (Ophir; Model 7Z01560; Tokyo, Japan). Laser-generated particle size, and the distribution and morphology of water-dispersed ceramic NPs were determined using scanning electron microscopy (SEM; JEOL JSM-5310, 15 kV; Tokyo, Japan) and transmission electron microscopy (TEM; JEOL JEM2100F, 200 kV; Tokyo, Japan). Crystal phase was also analyzed with an X-ray diffractometer
Fig. 2. (a) SEM micrograph at a water level of 2 mm, (b) and (c) TEM images at a water level of 4 and 6 mm, respectively; (d) histogram showing NP size distributions at 6 mm; laser power was set to 80 mW [fluence; 20 J/cm2].
48
S.A. Al-Mamun et al. / Thin Solid Films 523 (2012) 46–51
and laser fluence of 3.1 J/cm 2. In our experiment, the ablation rate was comparatively very high at a water level of only 2 mm, and it was found that irregular-shaped submicron- and micron-sized crater particles tend to dominate causing a high ablation weight, as shown in Fig. 2(a). This maximum ablation rate at a minimum water level can be attributed to fragmentation associated with melting of the target when the fluence is 20 J/cm 2. Fragmentation occurred due to lattice heat propagation, lattice expansion, thermally induced stress and brittle fracture [16], and shock-wave-induced mechanical vibration. At a water level of 4 mm, the ablation rate sharply decreased due to the absence of large irregular-shaped particles, indicating that mechanical stress was no longer present and all ablated particles were nano-sized and spherical resulting from melting, evaporation, and condensation. At higher water levels, the ablation rate gradually decreased; probably due to an increase in loss of incident power absorbed by the water and the already ablated particles present on the laser path. Thus, the optimum value of water layer thickness divides the water-level domain into two zones. Both fragmentation and melting occur at a water level below the optimum value, whereas only melting occurs above that value. Particle size (both d50 and average) tends to decrease with increasing water level, as shown in Fig. 3(b). The laser light scattering experiment conducted by Soliman et al. [17,18], and Sasaki and Takada [19] indicated the growth of particles inside the cavitation bubble, which actually contains the plume. They also found that the dynamics of the cavitation bubble, thus the size of the particles, could be controlled by water pressure. There is also a complex interaction among collision frequency of particles inside the plume; coalescence and shock wave intensity may define particle size at different water levels. The decrease in size at a higher water level could also be attributed to laser absorption by water, which could affect the plume temperature and collision frequency of particles. 3.2. Effect of laser power Fig. 3. (a) Dependence of ablation rate and (b) diameter of the synthesized NPs on a water level above the alumina target.
with CuKα1 radiation (RIGAKU Smartlab, 40 kV, 30 mA, wavelength 0.1544 nm; Tokyo, Japan). 3. Results and discussion Laser fluence, liquid level above the target, and laser spot size play an important role in determining the shape, size, morphology, phase, and ablation efficiency of the synthesized particles. 3.1. Effect of liquid level above target To investigate the effect of the liquid level on the particle size and productivity, the water level above the target was varied from 2 to 10 mm while keeping other parameters constant (power was kept at 80 mW, fluence; 20 J/cm2). Fig. 2(a) is an SEM micrograph of the ablated particles at a water level of 2 mm. Panels (b) and (c) of Fig. 2 are the TEM images of ablated NPs at a water level of 4 and 6 mm, respectively, while Fig. 2(d) shows the particle size distributions at 6 mm. The calculated mass-median-diameter (d50) was 28.5 nm and the arithmetic average of the NPs was 12.3 nm. ImageJ software was used and particle sizes ranging from 1 to 60 nm, which were clearly identified from the TEM micrographs, were taken into account when measuring NP size. Fig. 3(a) shows the ablation weight per hour at different water levels. According to Zhu et al. [9], ablation in a water-confined environment is greatly affected by the induced shock wave triggered by adiabatically expanding plasma; thus, the ablation rate should be maximum at the optimum water level. They found that the ablation rate was maximum at a water level of 1.1 mm by irradiating a Si target with a 248-nm KrF excimer laser with a pulse duration of 23 ns
Alumina, a wide band gap material, has a very narrow range of active laser fluence for synthesizing NPs. Laser power above the corresponding fluence causes mechanical stress; thus, large irregular-shaped cracked particles, while below that range, results in insufficient energy for ablation. For this reason, the choice of suitable fluence is an important factor. Panels (a) and (b) of Fig. 4 are the TEM images at laser powers of 80 and 120 mW. With a laser spot size of 250 μm, the calculated laser fluences at 80 and 120 mW were 20 and 30 J/cm2, respectively. In all the cases, the water level was kept to 4 mm. Fig. 4(c) is an SEM micrograph of the ablated particles at a laser power of 200 mW (fluence; 50 J/cm2). At laser fluences of 20 and 30 J/cm2, synthesized nanocondensates were less than 100 nm but at a fluence of 50 J/cm2, large spherical droplets of submicron- and even micron-sized particles were observed, which could be attributed to extreme coalescence and agglomeration of particles. Note that only spherical particles were observed even at high power, indicating that melting only occurs at this water level. Fig. 4(d) shows the variation in the ablation rate as a function of laser power at a water level of 4 mm. As expected, the ablation rate increased with power; thus, the laser fluence and particle size increased as suggested from the histograms in Fig. 5(a) and (b) for 80 and 120 mW, respectively. Fig. 5(c) also suggests that the particle diameter increases with incident laser power. At high power, temperature and pressure inside the plasma plume increase; this further intensifies the collision frequency among particles, leading to the coalescence and formation of large particles. 3.3. Crystal phase identification Fig. 6 shows X-ray diffraction patterns of (a) bulk α-Al2O3, and (b) Al2O3 NPs synthesized by laser ablation in water at a laser power of 80 mW (20 J/cm 2) and a water level of 4 mm, indicating
S.A. Al-Mamun et al. / Thin Solid Films 523 (2012) 46–51
49
Fig. 4. Electron micrographs (a), (b), and (c) at laser powers of 80, 120, and 200 mW, respectively, and at a water level of 4 mm, (d) dependence of ablation rate on laser power.
the mainly stable α-Al2O3 phase with one γ-Al2O3 peak at the (220) plane. It is evident that some laser generated NPs have identical trigonal Bravais lattice structures as the solid target, corresponding to the α-phase alumina with space group R3c, while others evaporated and nucleated particles agglomerated as the thermodynamically metastable γ-phase after solidification. Phase identification for individual particles was also done using high-resolution TEM (HRTEM) and fast Fourier transform (FFT) images. Small synthesized particles were spherical and single crystals with α or γ-phase, but relatively large particles were a result of coalescence of two or more small particles of the same phase showing multiple crystal planes. For example, panels (a) and (b) of Fig. 7 are the HRTEM, and FFT of the corresponding HRTEM images of a spherical particle of about 30 nm. The interplaner spacing (d-spacing) of various planes was calculated and compared with JCPDS, PDF2 database (release 2010), and the γ-phase was identified. HRTEM revealed a d-spacing of 0.46 and 0.24 nm for the corresponding crystal planes of (111) and (311), respectively, for γ-Al2O3. The FFT of the HRTEM image shows γ(111), γ(311), and γ(400) crystal planes. Panels (c) and (d) of Fig. 7 are the HRTEM and FFT images of another particle of almost the same size, which confirms α-Al2O3. The growth of the α(113) and α(104) planes with a tilting angle of 39.6° was observed on the FFT image. Determination of the crystal phase is a complex function of the particle cooling rate (especially in the liquid phase), specific surface area of NPs, and free energy of particles. From a thermodynamic point of view, the α-phase should be stable for the coarsely crystalline Al2O3. However, the γ-phase becomes more stable at the nanoscale due to its lower
surface energy; nevertheless, it exhibits higher entropy than α-Al2O3 [2]. McHale et al. [2] predicted that γ-Al2O3 should become an energetically stable crystal structure for a specific particle surface area exceeding 125 m2 g− 1 at room temperature. Sajti et al. [13] calculated the specific surface area with the following equation: ATot ¼ VMpAbl ⋅ρ Ap , where MAbl is the total ablated mass, ρis the density of the corundum, and Vp and Ap are the volume and effective surface area of a single spherical particle. They found stable α-phase for a particle size of 30 nm. According to this assumption, alumina NPs larger than 12 nm, with a density of 3.97 g/cm 3, should be of a stable α-phase. However, there has not been any strict boundary of the crystal phase of particles defined in terms of size. For example, Khan et al. [11] found their synthesized NPs to be γ-phase, even at a particle size of 29 nm. In our work, we found both α and γ-phases to be present at an average particle size between 12 and 18 nm. We did not find a very straight relationship between particle size and phase since both the stable α-phase and metastable γ-phase were observed for the same particle size. 4. Conclusion We investigated the dependence of synthesized NP productivity, size, and morphology on laser power and liquid level. The synthesized particles were spherical with an average size between 12 and 18 nm (d50; 26–34 nm) depending on the laser operating parameters. The formation mechanism involved evaporation, melting, and dissociation followed by chemical reactions, recombination and crystallization of NPs.
50
S.A. Al-Mamun et al. / Thin Solid Films 523 (2012) 46–51
(a)
(b)
(c)
Fig. 6. Crystal structure: X-ray diffraction patterns of (a) bulk α-Al2O3 target, and (b) Al2O3 NPs synthesized by laser ablation in water for a laser power of 80 mW and a water level of 4 mm, indicating a stable α-Al2O3 phase.
35 References
Diameter [nm]
30 d 50 25 20 Arithmetic average 15 10 80
90
100
110
120
Laser power [mW] Fig. 5. Histograms (a), and (b) showing NP size distributions at 80 and 120 mW, respectively, and (c) dependence of NP size on the incident laser power at a water level of 4 mm.
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19]
K. Fui, V. Wong, T. Kurma, Nanotechnology 19 (2008) 345702. J.M. McHale, A. Auroux, J. Perrotta, A. Navrotsky, Science 277 (1997) 788. D.J. Suh, T.-J. Park, J.-H. Kim, K.-L. Kim, Chem. Mater. 9 (1997) 1903. J.R. Jensen, T. Johannessen, S. Wedel, H. Livbjerg, J. Nanopart. Res. 2 (2000) 363. T. Johannessen, S.E. Pratsinis, H. Livbjerg, Chem. Eng. Sci. 55 (2000) 177. Y.Q. Wu, Y.F. Zhang, X.X. Huang, J.K. Guo, Ceram. Int. 27 (2001) 265. R. Sivarajan, S. Elena, C. Baruch, C. Rachman, G. Aharon, J. Am. Ceram. Soc. 83 (2000) 89. N. Bärsch, A. Gatti, S. Barcikowski, J. Laser Micro/Nanoeng. 4 (2007) 66. S. Zhu, Y.F. Lu, M.H. Hong, X.Y. Chen, J. Appl. Phys. 89 (2001) 2004. S. Zhu, Y.F. Lu, M.H. Hong, Appl. Phys. Lett. 79 (2001) 1396. S.Z. Khan, Z. Liu, L. Li, Appl. Phys. A 101 (2010) 781. C.L. Sajti, R. Sattari, B. Chichkov, S. Barcikowski, Appl. Phys. A 100 (2010) 203. C.L. Sajti, R. Sattari, B.N. Chikov, S. Barcikowski, J. Phys. Chem. C 114 (2010) 2421. I.L. Liu, P. Shen, S.Y. Chen, J. Phys. Chem. C 114 (2010) 7751. O.R. Musaev, A.E. Midgley, J.M. Wrobel, M.B. Kruger, Chem. Phys. Lett. 487 (2010) 81. R.L. Webb, L.C. Jensen, S.C. Langford, J.T. Dickinson, J. Appl. Phys. 74 (1993) 2338. W. Soliman, N. Takada, K. Sasaki, Appl. Phys. Express 3 (2010) 035201. W. Soliman, N. Takada, K. Sasaki, Jpn. J. Appl. Phys. 50 (2011) 108003. K. Sasaki, N. Takada, Pure Appl. Chem. 82 (2010) 1317.
S.A. Al-Mamun et al. / Thin Solid Films 523 (2012) 46–51
51
Fig. 7. HRTEM images (a), and (c) showing a crystallized particle; magnification of the marked area (inset) shows interplaner spacing, and (b), (d) are the corresponding FFT images showing spots matching γ-Al2O3 and α-Al2O3, respectively.