Read 05_APL_6.pdf text version

APPLIED PHYSICS LETTERS 87, 261913 2005

Etching silicon wafer without hydrofluoric acid

Hong Liua

State Key Laboratory of Crystal Materials, Shandong University, Jinan 250100 People's Republic of China and School of Materials Science and Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332-0245

Zhong Lin Wangb

School of Materials Science and Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332-0245

Received 11 July 2005; accepted 27 October 2005; published online 22 December 2005 A one-step hydrofluoric-acid-free hydrothermal-etching technique is demonstrated for the preparation of porous silicon with vertical holes. This method demonstrates a "green" chemical approach for etching a silicon wafer or the preparation of bismuth-silicon nanostructures without toxic acid or applying an external voltage. By controlling the heating temperature 180 ° C and time, nanoscale vertically holed porous silicon has been created. A formation mechanism has been proposed on the basis of experimental observations. © 2005 American Institute of Physics. DOI: 10.1063/1.2158021 Porous silicon has attracted growing interest because of its potential applications in optoelectronics, microelectronics, photoelectrochemical solar cells, chemical and biochemical sensors, and drug delivery based on its photoluminescence properties, semiconductor properties, large surface area, and strong absorption ability.1­4 Normally, porous silicon is prepared by the electrochemical anodization etching technique in a hydrofluoric acid based solution. However, the setup and the number of process parameters of this technique are complex, resulting in high cost for preparation of porous silicon. Moreover, hydrofluoric acid is corrosive and hypertoxic, which is harmful not only for the operators but also for the environment. Recently, a new metal-assistanted etching technique is of interest because its processing simplicity and good photoluminescence properties.5,6 The improved method uses no bias voltage and is suitable for etching undoped silicon wafers. However, the toxic HF is still unavoidable for etching. To establish a hydrofluoric acid-free etching technique, an electrochemical etching method with hydroxide solution as the corrodant has been applied for preparation of porous silicon.7 Unfortunately, the etching process has to be stopped and restarted after a certain time, because a silicon oxide layer is forming at the bottom of the etched holes when the time of the applying voltage on the silicon wafer is long; the oxide layer has to be removed by hydrofluoric acid,8,9 otherwise, the etching process cannot continue. So far, no technique is available for etching a silicon wafer without using either hydrofluoric acid or an applied voltage. Therefore, finding a "green" etching method without hydrofluoric acid becomes a great challenge for the application of silicon. Here, we report a hydrothermal method for etching an undoped silicon without HF and an applied voltage. The corrodent is only nontoxic bismuth hydroxide. During the hydrothermal process, vertical etched holes form on surfaces of the silicon wafer, and at the same time, bismuth macroballs or nanospheres grow in situ at the bottom of the etched holes. This green chemical technique could be important for

a b

Electronic mail: [email protected] Electronic mail: [email protected]

preparing porous silicon based nanostructures. The etching process was performed in an autoclave. The preparation process comprised three steps: 1 cleaning the silicon wafers undoped 001 silicon wafer sequentially using H2SO4 and H2O2 H2SO4 97% / H2O2 30% = 7 / 1 , 90 ° C 30 min , deionized water ultrasonic bath 10 min, 4­6 times, to pH = 7 , and ethanol ultrasonic bath 10 min . 2 Preparing 0.01 mol/ L bismuth hydroxide suspension: 0.0005 mol of analytical grade Bi NO3 3 · 5H2O 98%, Alfa was dissolved in 50 ml deionized water to get bismuth nitric solution. Excessive ammonia hydroxide solution was dropped into bismuth nitric solution to get bismuth hydroxide precipitate. The precipitate was filtrated and washed by deionized water several times until the pH value reached 7. Then, 0.01 mol/ L bismuth hydroxide suspension was prepared by dispersing all of the precipitate in deionized water under ultrasonic irradiation for 60 min and adding deionized water to 50 ml. 3 Hydrothermal process: The earlier silicon wafer was placed at the bottom of a 22 ml a teflonlined stainless steel autoclave. 2 ml of the bismuth hydroxide suspension and 18 ml of deionized water were added into the autoclave. The autoclave was sealed and put into a furnace, which was preheated to 160­ 180 ° C. After heated for a different time, the autoclave was taken out and cooled down to room temperature. The silicon wafer was taken out from the autoclave, washed by deionized water and then by alcohol, and dried by flowing air. X-ray diffraction XRD measurements demonstrated that the samples synthesis at 160­ 175 ° C for 15­ 30 hours are composed of hexagonal bismuth JCPD-85-1330 and single crystal silicon with diamond structure JCPD-800018 . However, for the samples reacted for 48 h or longer, beside the Si 400 peak, we cannot find any peak of hexagonal bismuth or other crystalline phases in the XRD pattern. Scanning electron microscopy SEM was used to check the morphology of the sample. Figure 1 shows SEM images of the sample etched at 175 ° C at different times. Figures 1 a and 1 b are SEM images of the top view and fracture surface of a commercial silicon wafer, respectively. The wafer is smooth, and no obvious defect can be found on the

0003-6951/2005/87 26 /261913/3/$22.50 87, 261913-1 © 2005 American Institute of Physics Downloaded 04 Jan 2006 to 130.207.165.29. Redistribution subject to AIP license or copyright, see http://apl.aip.org/apl/copyright.jsp

261913-2

H. Liu and Z. L. Wang

Appl. Phys. Lett. 87, 261913 2005

FIG. 1. SEM images of the Si 001 sample etched at 175 ° C for different times. a , b Top view and fracture surface morphology of commercial 001 silicon wafer before etching; c , d top view and fracture section of a sample reacted at 175 ° C for 15 h, and EDS of an individual ball inset ; e , f top view and fracture surface of a sample reacted at 175 ° C for 24 h and g for 30 h; h the surface of the wafer etched at 175 ° C for 48 h; i SEM images acquired from a fracture section; j , k cylinders at the bottom and at the middle of the etching path, and the EDS from a cylinder inset .

surface. From Figs. 1 c and 1 d , after reacting at 175 ° C for 15 h, there are some small balls of 1.5­ 2 m in diameter distributed homogeneously on the surface of the silicon wafer. Energy dispersed x-ray spectroscopy EDS analysis indicates that the ball is elemental bismuth, which is consistent with the XRD data. The bismuth balls sank into etched holes and formed a ball-hole structure in the Si wafer reacted for 24 h Figs. 1 e and 1 f . For the sample reacted for 30 h, the etching holes became deeper and the balls sank into the holes completely Fig. 1 g . At last, when the reaction time is for 48 h and longer, the etched holes became very deep, and the surface of the wafer was covered by a layer of thin flakes Fig. 1 h ; EDS indicates the composition of the surface contains only oxygen and silicon. We have demonstrated by XRD that there is no crystalline substance besides

Downloaded 04 Jan 2006 to 130.207.165.29. Redistribution subject to AIP license or copyright, see http://apl.aip.org/apl/copyright.jsp

single crystalline silicon for the samples reacted at 175 ° C for over 48 h. Therefore, the nanoflakes on the surface of the wafer should be amorphous silicon oxide. Figure 1 i is the fracture section of the wafer etched for 48 h. The depth of the holes is about 5 m. From this figure, we can see that the diameter of the holes decreases with increase of the hole depth. In the fracture section of the etched wafer, a small cylinder was formed at the bottom of each hole. Figure 1 j shows the cylinders near the bottom of the holes. Their diameters are about 100 nm, much smaller than that of the balls at the surface of the wafer etched for 15 or 24 h. Figure 1 k is the morphology of a cylinder located at the middle position between the opening and the bottom of a hole, and this configuration is dominant. Their diameters are about 200 nm, and are much larger than that of the cylinders at the bottom of the holes. The inset in Fig. 1 k is an EDS acquired from the cylinder. It proves that the cylinder should be bismuth. The results indicate that silicon wafer can be etched by bismuth hydroxide, and the etched holes deepen during the etching process. The bismuth balls formed during the etching process should play an important role in etching. The initial size of bismuth ball should decide the diameter of the hole. To receive nanosize etching holes on the silicon wafer, samples were prepared at a lower temperature for different times. Figures 2 a and 2 b are top view and fractured section of the samples prepared at 160 ° C for 15 h. Small spheres 150­ 200 nm in diameters can be found on the surface of the silicon wafer. However, no obvious etched hole can be found. With an increase of reaction time, the silicon wafer was etched and the balls sank into the etched holes. The etching process is the same as that at 175 ° C Figs. 2 c ­2 e . When the reaction time reaches 48 h or longer, no nanosphere appeared on the surface. There are a lot of opening holes on the surface of the wafer Fig. 2 f . From the image of the fraction section of the wafer, the depth of etching is more than 30 m Fig. 2 g , which is much deeper than that of the wafer etched at 175 ° C for the same time. Small nanodisks of 50­ 10 nm in diameter can be found at the bottom of the holes. We can get uniform porous silicon structure by optimization of the etching condition at 160 ° C for 72 h Fig. 2 i . Based on the experimental results, an etching mechanism is proposed. Silicon atoms at the surface of the silicon wafer can react with hydroxyl anions decompounded from bismuth hydroxide and dissolve in the solution and produce some elemental hydrogen atoms. Therefore, the silicon wafer is etched and forms some shallow pits. At the same time, the bismuth ions in the solution can be reduced by hydrogen atoms and accumulate to be crystalline nuclei Fig. 3 a and then form crystalline Bi spheres located at the pits Fig. 3 b . The bismuth balls play a role as catalyst and "dig" the hole by "catching" silicon atoms at the bottom of the holes to form unstable temporary alloy cluster and the "releasing" then to the solution. At a primary stage of the etching process, the bismuth balls in the hole keep growing, and the diameter of the hole should increase, because the concentration of bismuth ions is high and the ball is exposed to the solution directly Fig. 3 c . After the balls totally sink into the hole, they will stop growing, although the depth of the hole increases instantly because of lacking of bismuth ions in the hole.

261913-3

H. Liu and Z. L. Wang

Appl. Phys. Lett. 87, 261913 2005

FIG. 3. A schematic etching process and the evidence for each etching step: a formation of a shallow pit and a nuclei of bismuth cluster; b the formation of a bismuth ball and its beginning to sink into the etched hole; c the bismuth ball sinks totally into the hole; d the hole is deepened and the ball deforms as etching proceeds at higher temperature; and e the change of the hole and the ball at lower temperature.

consumption of the bismuth element during the catalysis process. In conclusion, we have developed a one-step hydrofluoric-acid-free hydrothermal-etching technique for the preparation of porous silicon with vertical holes. This technique also produces bismuth spheres at the bottom of the holes. By controlling the heating temperature and time, nanoscale vertically holed porous silicon have been created. This method demonstrates a green chemical approach for etching a silicon wafer or the preparation of bismuth-silicon nanostructures without using toxic chemicals or applying external voltage. A formation mechanism has been proposed on the basis of experimental observations. This process is likely to have important applications for preparing silicon based micro- and nanostructures on a large scale.

FIG. 2. SEM images of the samples etched at 160 ° C for different time. a From fractured side of the wafer; b top view of the wafer after etching for 15 h; c top view of the wafer etched for 24 h, d top view of the wafer etched for 30 h, e from the fractured side of the same sample, f , g top view and fractured side surface of a wafer etched for 48 h; h high resolution image of g ; and i from the fractured side of the sample synthesized at an optimized etching condition of 160 ° C for 72 h.

The authors are thankful for the support from the National High Technology Research and Development Program of China 863 Program, No. 2002AA31110 , NSFC 50572052 , NSF DMI 0403671 , the NASA Vehicle Systems Program and Department of Defense Research and Engineering DDR&E , and the Defense Advanced Research Projects Agency Award No. N66001-04-1-8903 .

R. P. Chin, Y. R. Shen, and V. Petrova-Koch, Science 270, 776 1995 . V. S.-Y. Lin, K. Motesharei, K.-P. S. Dancil, M. J. Sailor, and M. Reza Ghadiri, Science 278, 840 1997 . 3 Y. Y. Li, F. Cunin, J. R. Link, T. Gao, R. E. Betts, S.-H. Reiver, V. Chin, S. N. Bhatia, and M. J. Sailor, Science 299, 2045 2003 . 4 S. Yae, Y. Kawamoto, H. Tanaka, N. Fukumuro, and H. Matsuda, Electrochem. Commun. 5, 632 2003 . 5 X. Li and P. W. Bohna, Appl. Phys. Lett. 77, 2572 2000 . 6 K. Tsujino and M. Matsumura, Adv. Mater. Weinheim, Ger. 17, 1045 2005 . 7 D. L. Kendall, Appl. Phys. Lett. 26, 195 1975 . 8 O. J. Glembocki, R. E. Stahlbush, and M. J. Tomkiewicz, J. Electrochem. Soc. 132, 145 1985 . 9 P. Raisch, W. Haiss, R. J. Nichols, and D. J. Schiffrin, J. Phys. Chem. B 105, 12508 2001 . 10 Y. W. Wang, B. H. Hong, and K. S. Kim, J. Phys. Chem. B 109, 7068 2005 .

2 1

The catalyst bismuth ball in the hole changes its shape and volume during the etching process. Only the atoms at the most frontier part of the ball easily "catch" and "release" silicon atoms. When bismuth atoms release from the temporary alloy clusters, they easily transfer and are deposited on the back of bismuth ball surface, which cause the shape of the ball change. At a higher etching temperature, because the bismuth ball is very large, and only the most frontal part of the ball can help to etch the hole, the diameter of the hole will become smaller. Therefore, the ball will become cylindrical to adapt the small diameter of the hole Fig. 3 d . At a lower etching temperature, because the bismuth nanosphere is very small and the etching speed is very high, the sphere is easily to keep the spherical shape. However, the sphere shrinks, and become a disk Fig. 3 e , because the

Downloaded 04 Jan 2006 to 130.207.165.29. Redistribution subject to AIP license or copyright, see http://apl.aip.org/apl/copyright.jsp

Information

3 pages

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate

997198


Notice: fwrite(): send of 196 bytes failed with errno=32 Broken pipe in /home/readbag.com/web/sphinxapi.php on line 531