Thermal Behavior and Film Formation from an Organogermanium

Hsiang Wei Chiu, and Susan M. Kauzlarich*. Department of Chemistry, One Shields Avenue, University of California Davis, Davis, California 95616. Eli S...
0 downloads 0 Views 1MB Size
Langmuir 2006, 22, 5455-5458

5455

Thermal Behavior and Film Formation from an Organogermanium Polymer/Nanoparticle Precursor Hsiang Wei Chiu and Susan M. Kauzlarich* Department of Chemistry, One Shields AVenue, UniVersity of California DaVis, DaVis, California 95616

Eli Sutter* Center for Functional Nanomaterials, BrookhaVen National Laboratory, Upton, New York 11973 ReceiVed December 9, 2005. In Final Form: January 31, 2006 In situ high-resolution transmission electron microscopy (HRTEM) was used to investigate the effect of heating on an organo-Ge polymer/nanoparticle composite material containing 4-8 nm diameter alkyl-terminated Ge nanoparticles. The product was obtained from the reduction of GeCl4 with Na(naphthalide) with subsequent capping of the -Cl surface with n-butyl Grignard reagent. The in situ HRTEM micrographs show that the product undergoes significant changes upon heating from room temperature to 600 °C. Two pronounced effects were observed: (i) Ge nanoparticles coalesce and remain crystalline throughout the entire temperature range, and (ii) the organo-Ge polymer acts as a source for the in situ formation of additional Ge nanoparticles. The in situ-formed Ge nanoparticles are approximately 2-3 nm in diameter. These in situ-formed nanoparticles (2-3 nm) are so dense that, together with the original ones, they build up an almost continuous crystalline film in the temperatures between 300 and 500 °C. Above 480 °C, melting of the in situ formed Ge nanoparticles (2-3 nm) is observed, while nanoparticles greater than 5 nm remain crystalline. After cooling to room temperature, the 2-3 nm Ge nanoparticles recrystallized.

Introduction Recently, the synthesis of Ge nanoparticles (NPs) has received considerable attention, with several groups making great strides with size, shape, and surface control.1-6 Germanium NPs are of interest for many reasons, such as the study and application of expected quantum confinement effects.7-10 Here, we are interested in using an organic soluble form of Ge NPs for the production of large-area Ge films. Ge films form the transition layer in the high-energy, lightweight GaAs-based photovoltaics.9,11,12 Ge is the material of choice since crystalline Ge is lattice matched to GaAs and is suitable as a substrate for GaAs-based multijunction photovoltaics. The energy conversion of triple junction Ge/GaAs photovoltaics is about 26%, which is the highest currently known.11 Ge/GaAs heterostructured films are typically prepared via epitaxial growth techniques such as molecular beam epitaxy (MBE) and metal organic vapor phase epitaxy (MOVPE).13 There have been efforts in producing films from liquid-phase processes14-16 to reduce costs and facilitate large-area device fabrication. Films of Ge have been recently prepared from * To whom correspondence should be addressed. [email protected] (S.M.K.); [email protected] (E.S.).

E-mail:

(1) Taylor, B. R.; Fox, G. A.; Hope-Weeks, L. J.; Maxwell, R. S.; Kauzlarich, S. M.; Lee, H. W. H. Mater. Sci. Eng., B 2002, 96, 90-93. (2) Hope-Weeks, L. J. Chem. Commun. 2003, 2980-2981. (3) Gerion, D.; Zaitseva, N.; Saw, C.; Casula, M. F.; Fakra, S.; van Buuren, T.; Galli, G. Nano Lett. 2004, 4, 597-602. (4) Lu, X.; Ziegler, K. J.; Ghezelbash, A.; Johnston, K. P.; Korgel, B. A. Nano Lett. 2004, 4, 969-974. (5) Fok, E.; Shih, M.; Meldrum, A.; Veinot, G. C. Chem. Commun. 2004, 386-387. (6) Lu, X.; Korgel, B. A.; Johnsson, K. P. Chem. Mater. 2005, 17, 6479-6485. (7) Bostedt, C.; van Buuren, T.; Willey, T. M.; Franco, N.; Terminello, L. J.; Heske, C.; Moller, T. Appl. Phys. Lett. 2004, 84, 4056-4058. (8) Chang, T. C.; Yan, S. T.; Liu, P. T.; Chen, C. W.; Lin, S. H.; Sze, S. M. Electrochem. Solid-State Lett. 2004, 7, G17-G19. (9) Mauk, M. G.; Balliet, J. R.; Feyock, B. W. J. Cryst. Growth 2003, 250, 50-56. (10) Teo, L. W.; Choi, W. K.; Chim, W. K.; Ho, V.; Moey, C. M.; Tay, M. S.; Heng, C. L.; Lei, Y.; Antoniadis, D. A.; Fitzgerald, E. A. Appl. Phys. Lett. 2002, 81, 3639-3641. (11) Blondeel, A.; Clauws, P.; Depuydt, B. Mater. Sci. Semicond. Process. 2001, 4, 301-303. (12) Bailey, S. G.; Flood, D. J. Prog. PhotoVoltaics: Res. Appl. 1998, 6, 1-14.

organosoluble germanium nanoclusters (OGEs) and have been shown to be capable of signal amplification as a field-effect transistor.14,17 Heat treatment of the deposited amorphous film shows crystallization initiates at 500 °C, which has been verified by Raman spectroscopy and electrical conductivity.17 Size-dependent melting of nanocrystals has been studied for many years, both experimentally and theoretically. In 1954, Takagi18 experimentally revealed that the melting point depression of metallic nanocrystals corresponded to their bulk melting temperatures through transmission electron microscopy (TEM). It is now known that the melting temperature of all lowdimensional crystals, including metallic,19-23 organic,24,25 and semiconductor,26,27 depends on their size. For free-standing nanocrystals, the melting temperature decreases as its size decreases.18-28 It has been shown, in the case of Si NPs that the melting point (Tm) of the smallest nanocluster was depressed by a factor of 3.26,27 Therefore, a similar trend in the depression of the melting point of germanium is anticipated. TEM has been widely used (13) Usami, N.; Azuma, Y.; Ujihara, T.; Sazaki, G.; Nakajima, K.; Yakabe, Y.; Kondo, T.; Kawaguchi, K.; Koh, S.; Shiraki, Y.; Zhang, B. P.; Segawa, Y.; Kodama, S. Semicond. Sci. Technol. 2001, 16, 699-703. (14) Watanabe, A.; Unno, M.; Hojo, F.; Miwa, T. J. Mater. Sci. Lett. 2001, 20, 491-493. (15) Watanabe, A.; Unno, M.; Hojo, F.; Miwa, T. Mater. Lett. 2001, 47, 8994. (16) Watanabe, A.; Unno, M.; Hojo, F.; Miwa, T. Mater. Lett. 2002, 4286, 1-5. (17) Watanabe, A.; Hojo, F.; Miwa, T. Appl. Organomet. Chem. 2005, 19, 530-537. (18) Takagi, M. J. Phys. Soc. Jpn. 1954, 9, 359-363. (19) Hasegawa, M.; Hoshino, K.; Watabe, M. J. Phys. F: Met. Phys. 1980, 10, 619-635. (20) Jiang, Q.; Aya, N.; Shi, F. G. Appl. Phys. A: Mater. Sci. Process. 1997, 64, 627-629. (21) Jiang, Q.; Tong, H. Y.; Hsu, D. T.; Okuyama, K.; Shi, F. G. Thin Solid Films 1998, 312, 357-361. (22) Peters, K. F.; Cohen, J. B.; Chung, Y. W. Phys. ReV. B 1998, 57, 1343013438. (23) Dippel, M.; Maier, A.; Gimple, V.; Wider, H.; Evenson, W. E.; Rasera, R. L.; Schatz, G. Phys. ReV. Lett. 2001, 87, 095505-095501. (24) Jiang, Q.; Zhang, Z.; Hsu, D. T.; Tong, H. Y.; Iskandar, M. J. Mater. Sci. 1999, 34, 5919-5922. (25) Morishige, K.; Kawano, K. J. Phys. Chem. B 1999, 103, 7906-7910.

10.1021/la053343p CCC: $33.50 © 2006 American Chemical Society Published on Web 05/05/2006

5456 Langmuir, Vol. 22, No. 12, 2006

Chiu et al.

to study size-dependent melting point depression.26,27,29,30 This paper focuses on the investigation of the thermal behavior of an organo-Ge polymer/NP composite in the temperature range from room temperature to 600 °C using an in situ high-resolution transmission electron microscope (HRTEM). This compliments our previous publication31 in which we used differential scanning calorimetry (DSC) to measure the melting of Ge NP powders. We presented the reaction of GeCl4 with sodium naphthalide, with subsequent -Cl replacement with n-butyl groups, and demonstrated that the product was an organo-Ge polymer/NP composite.31 We showed that these Ge NPs, prepared via sodium naphthalide reduction, were initially crystalline, and that, with removal en vacuo of surface ligands and the organo-Ge polymer, they became amorphous, as determined by selected area electron diffraction (SAED).31 These amorphous NPs recrystallized at 561 °C, grew to about 8 nm in diameter and subsequently sintered and melted at 925 °C. This paper presents the in situ HRTEM characterization of the organo-Ge polymer/NP composite prepared via reduction with sodium naphthalide. The in situ HRTEM provides a method for investigating the behavior of individual NPs as a function of temperature. This study reveals the formation of a nearly continuous Ge film due to the contribution of the organo-Gecontaining polymer to the NP population with increasing temperature. Experimental Section Materials. Ethylene glycol dimethyl ether (glyme) (Acros, 99+%) was dried and distilled twice from a sodium-potassium (Na-K) alloy under argon. The Na-K alloy was freshly prepared from a mixture of sodium (Aldrich, 99%) and potassium pieces (Aldrich, 98%). Naphthalene, (C10H8) (Fisher, refined, 99.98%), germanium tetrachloride, (GeCl4) (Acros, 99.99%), and butylmagnesium chloride (Aldrich, 2 M) were used without further purification. HPLC-grade water (EM Science) and HPLC-grade hexane (Aldrich) were used as received. Manipulations of these chemicals were handled either in a nitrogen-filled glovebox or on a Schlenk line using standard anaerobic and anhydrous techniques. Synthesis of Organogermanium Polymer/NP Composite. Na metal (0.5190 g, 22.58 mmol) was added to a Schlenk flask in a drybox and transferred to a Schlenk line. Then 2.894 g (22.58 mmol) of naphthalene was added under argon. Approximately 70 mL of freshly distilled and degassed glyme was added to the solids, and the mixture was stirred overnight. Upon dissolution of the Na metal, the solution changed from being colorless to having a dark green color. A 70 mL solution of Na(naphth) was added rapidly via cannula to 0.70 mL (6.02 mmol) of GeCl4 in 300 mL of glyme in a Schlenk flask at room temperature. The solution immediately changed from a clear to a black suspension upon the addition of the Na(naphth) mixture. After 10 min of stirring, the suspension was allowed to settle. Once settled, there was a black solid at the bottom of the flask and a dark yellow solution on the top. The orange solution was vacuum-dried to remove the naphthalene. Freshly distilled and degassed glyme (250 mL) was then added to the flask, followed by 3.01 mL (6.02 mmol) of n-BuMgCl. The mixture was left to stir for 12 h. The mixture was pumped down to dryness, the organicterminated product was extracted with hexane, and the extract was rinsed with acidified water. After the removal of hexane in vacuo, with mild heating, ∼500 mg of viscous orange oil was obtained. Characterization. TEM and energy-dispersive X-ray (EDX) spectroscopy analyses of these NPs were performed on a Philips (26) Goldstein, A. N.; Echer, C. M.; Alivisatos, A. P. Science (USA) 1992, 256, 1425-1427. (27) Goldstein, A. N. Appl. Phys. A: Mater. Sci. Process. 1996, 62, 33-37. (28) Jiang, Q.; Shi, H. X.; Zhao, M. J. Chem. Phys. 1999, 111, 2176-2180. (29) Buffat, P.; Borel, J.-P. Phys. ReV. A 1976, 13, 2287-2298. (30) Castro, T.; Reifenberger, R. Phys. ReV. B 1990, 42, 8548-8556. (31) Chiu, H. W.; Chervin, C. N.; Kauzlarich, S. M. Chem. Mater. 2005, 17, 4858-4864.

Figure 1. (A) TEM micrograph that shows an ensemble of the NPs at room temperature. (B) HRTEM micrograph of Ge NPs. (C) HRTEM micrograph of a spherical shaped single Ge NP showing lattice fringes. CM-12, operating at 100 keV. TEM samples were prepared by dipping the holey, carbon-coated, 400-mesh electron microscope grid into the hexane solution of the orange oil, followed by heating in a 120 °C oven overnight. The NPs have also been characterized by chemical analysis, as described in previous publications.31,32 In Situ HRTEM. In situ experiments were carried out in a JEOL JEM 3000F field-emission TEM operating at 300 keV using a Gatan 652 high-temperature sample holder. The experiments are carried out in the temperature range between room-temperature and 600 °C at pressures below 2 × 10-5 Pa, and at electron irradiation intensities ∼2 A/cm2 to prevent any electron-beam-induced structural changes in the NPs. Micrographs were collected at room temperature (160 °C) and then in 10° increments to 200 °C, 20° increments to 300 °C, and 30° increments to 600 °C.

Results and Discussion The organo-Ge polymer/NP composite was synthesized by the reduction of germanium tetrachloride with sodium naphthalide, and the -Cl groups were terminated with n-butyl following a synthetic route described previously.31-33 We have shown that the product is a composite of organo-Ge polymer and Ge NPs with a combination of TEM, chemical analysis, and en vacuo heating experiments.31 The n-butyl-terminated Ge NPs were characterized by TEM as shown in Figure 1A. This sample showed a fairly large size distribution. The average diameter of NPs prepared for this study was 7.2 nm. We have already presented a detailed analysis on the reaction conditions and how (32) Chiu, H. W.; Kauzlarich, S. M. Chem. Mater. 2006, 18, 1023-1028. (33) Baldwin, R. K.; Pettigrew, K. A.; Ratai, E.; Augustine, M. P.; Kauzlarich, S. M. Chem. Commun. 2002, 1822-1823.

HRTEM Analysis of Organo-Ge Polymer/NP Composites

Figure 2. (A) HRTEM micrograph showing Ge NPs at 160 °C. (B) HRTEM micrograph of one Ge NP showing 2.08 Å lattice fringes. (C) HRTEM micrograph showing Ge NPs at 170 °C. (D) HRTEM micrograph of NPs showing stacking faults formation.

they affect the initial size of NPs prepared via this route.31,32 Figure 1B shows an initial ensemble of Ge NPs at roomtemperature with higher magnification. The particles are crystalline, and the lattice fringes with spacings of 2.00 and 1.65 Å, corresponding to Ge {220} and {311} planes, respectively, can be resolved. The lack of lattice fringes in all the NPs is due to the fact that they are not all fulfilling the Bragg condition at the same time. Lattice fringes can be observed in all particles during tilting. At each temperature, a large number of images were taken and analysis of the NPs crystallinity was performed. Figure 2A shows the bright-field micrograph of Ge NPs after heating to 160 °C. We started this experiment at 160 °C because the grids are dried overnight at 120 °C to remove any remaining solvent. At this temperature, the NPs showed faceting and coalescence. NPs size analysis shows that they become slightly larger, and some have coalesced. Figure 2B shows a highresolution image of a faceted Ge NP. The lattice fringes of 1.65 Å can be observed, which are indexed to Ge {311}. Figure 2C is the bright-field micrograph taken at 170 °C showing that already

Langmuir, Vol. 22, No. 12, 2006 5457

at this temperature some coalescence events are taking place. Figure 2D shows a high-resolution image of two particles that have coalesced. As a result, they are replaced by a new larger particle that shows a clear stacking fault. At temperatures above 300 °C, two trends are clearly observed in the experiment. The coalescence of the original Ge NPs continues with some of the coalesced particles now larger than 20 nm (Figure 3). The second process that takes places is associated with the presence of the organo-Ge polymer together with the original Ge NPs. We previously reported that this particular synthetic route to Ge NPs produces both an organo-Ge polymer and Ge NPs.31 In this experiment, we speculate that this polymer acts as a source of Ge and leadsto the in situ formation of small Ge NPs with sizes on the order of 2-3 nm in diameter. The size of these new NPs will depend on the amount of polymer as well as the annealing time. The newly formed NPs are crystalline and show lattice fringes. A variety of lattice spacings were determined that correspond to the diamond Ge phase {400}, {311}, and {220} (PDF no. 04-0545). However, some NPs show lattice spacing of 2.20 Å that is attributed to tetragonal Ge {103} (PDF no. 18-0549). Therefore, the newly formed NPs may be a mixture of cubic and tetragonal Ge. The tetragonal form of Ge is a metastable, high-pressure form of the structure. It has been obtained in small Ge nanocrystals deposited by a cluster-beam evaporation technique.34 It is postulated that the small size favors the more compact, high-pressure structure. In addition, in the in situ HRTEM experiment described herein, the additional strong surface tension due to the removal of organic surface groups may be expected to favor this structure. It has also been shown that, upon heating, this tetragonal structure will transform to cubic germanium.34 The density of the newly formed NPs is high and contributes to the formation of an almost continuous Ge film made of NPs. Figure 4A,B shows high-resolution images at various temperatures between 400 and 500 °C of the almost continuous film made of the original Ge particles and the newly formed ones.It should be noted that, at these temperatures, all the NPs remain crystalline and show lattice fringes. At temperatures above 480°C (Figure 4C) the smaller NPs, that is, those with diameters of less than 5 nm, no longer show lattice fringes. This is attributed to the melting of particles of this size range. As expected, the larger NPs remained crystalline. Figure 4D is a high-resolution micrograph of Ge NPs heated to 590 °C. Figure 5A shows the bright-field micrograph of Ge NPs taken after heating to 600 °C and cooling to room temperature. The NPs are clearly larger than they were before the heating experiment. In addition, there are a large number of small NPs that are now apparent that were formed during the experiment. Figure 5B shows a high-resolution image of one of the crystalline NPs. A comparison between Figures 5A and 1A clearly shows the increase in number of NPs that occurs during the experiment.

Summary A composite of n-butyl-terminated Ge NPs and polymer have been investigated by in situ HRTEM. The organo-Ge polymer component acts as a source for the formation of small Ge NPs at temperatures above 300 °C. The newly formed small Ge NPs, together with the original ones, contributed to the buildup of an almost continuous Ge film at temperatures above 400420 °C. The newly formed Ge NPs with sizes on the order of 2-3 nm became amorphous around 480 to 500 °C. After (34) Sato, S.; Nozaki, S.; Morisaki, H. Appl. Phys. Lett. 1998, 72, 2460-2462.

5458 Langmuir, Vol. 22, No. 12, 2006

Chiu et al.

Figure 4. (A) HRTEM micrograph showing Ge NPs at 420 °C. (B) HRTEM micrograph showing Ge NPs at 450 °C. (C) HRTEM micrograph showing Ge NPs at 500 °C. (D) HRTEM micrograph of Ge NPs at 590 °C.

Figure 5. (A) HRTEM micrograph showing Ge NPs that have been heated to 600 °C and cooled to room temperature. (B) HRTEM micrograph of a spherical-shaped Ge NP showing lattice fringes of 2.12 Å.

heating range. These studies suggest that film formation from NP precursors benefits from the presence of the organo-Ge polymer, which may impart additional solubility and homogeneity. Figure 3. (A) HRTEM micrograph showing Ge NPs at 300 °C. (B) HRTEM micrograph showing Ge NPs at 330 °C; the NPs are forming a quasi-continuous film. (C) HRTEM micrograph showing Ge NPs at 390 °C.

quenching the sample from 600 °C to room temperature, these small particles recrystallized. The larger particles, that is, those 5 nm diameter and larger, remained crystalline over the entire

Acknowledgment. We thank Donna Senft and Rajiv J. Berry for useful discussion. Funding from AFOSR, NSF DMR0120990, and CHE-0304871 is gratefully acknowledged. This work was performed in part under the auspices of the U.S. Department of Energy, under Contract No. DE-AC0298CH1-886. LA053343P