JP7108160B2 - Method for manufacturing silicon quantum dots - Google Patents

Description

The present invention relates to a method for manufacturing silicon quantum dots.

In recent years, semiconductor nanocrystal particles (quantum dots) have attracted attention as light-emitting materials for LEDs, displays, bio-imaging, etc., and several companies (NN-LABS, Nanoco, Nanosys, PlasmaChem, etc.) have been conducting research and development. ing. Semiconductor nanocrystals are often produced as colloids dispersed in a liquid because they are supplied in a sheet form for displays, for example. However, currently commercially available semiconductor quantum dots are compound semiconductors and often contain harmful metals such as cadmium and lead. Another issue is the improvement of raw material utilization efficiency and process energy utilization efficiency. In other words, for the production of semiconductor quantum dots, a high-luminescence performance, low-cost, mass production method using materials with high environmental compatibility (cadmium alternative) as raw materials is required.

Silicon is the most representative semiconductor and an indispensable base material for the current electronics industry. However, although silicon is a good electronic material, it has been considered difficult to apply it to light-emitting materials because of its properties as an indirect bandgap semiconductor. However, since the report in 1990 of visible light emission due to the quantum size effect from porous nanocrystalline silicon (single crystal with a diameter of about several nanometers) (Non-Patent Document 1), application of silicon to light-emitting materials has begun. Active research is underway. Since silicon exists abundantly in the earth's crust and is inexpensive, nanocrystalline silicon that emits visible light can be used as a substitute for scarce materials such as gallium-based compounds used in light-emitting diodes. big. In recent years, the use of highly dispersible compound-based semiconductor nanocrystal particles (quantum dots) in fluorescent labeling for bioimaging has been investigated because of their high emission wavelength controllability and luminous efficiency (non-quantum dots). According to Patent Document 2), these compound nanoparticles contain materials such as cadmium that are harmful to living organisms. On the other hand, since nanocrystalline silicon is harmless to the human body, it can be expected to be suitable for biological applications by making it into quantum dots (a particle shape that can be dispersed in a solution).

Typical methods for producing (manufacturing) silicon quantum dots include liquid-phase chemical synthesis (Non-Patent Document 3), chemical methods such as chemical etching (Patent Document 1), ion implantation (Non-Patent Document 4), and sputtering. physical methods such as the method (Non-Patent Document 5). Due to the quantum size effect, quantum dots exhibit a size-dependent change in bandgap and luminescence transition probability (the number of light emissions per unit time). ). In other words, the emission wavelength and emission efficiency can be controlled by the particle size. Moreover, not only the particle size but also the surface modification (terminal state) greatly affects the luminescent physical properties (bandgap energy and luminescent transition probability) of quantum dots. For example, in hydrogen-terminated or oxygen-terminated quantum dots, the quasi-direct transition makes the emission transition lifetime (reciprocal of the transition probability) from several microseconds to several hundreds of microseconds, and the emission wavelength ranges from orange to the near-infrared region (non Patent document 6). On the other hand, in a carbon-terminated quantum dot terminated with an organic molecule such as a hydrocarbon group, the electronic state changes compared to a normal silicon crystal, and the transition probability increases (Non-Patent Document 7). Therefore, carbon-terminated silicon quantum dots are the most promising system for light-emitting device applications.

In the production of silicon quantum dots, control of both particle size and surface termination states are usually important. As for the surface state, if no special surface modification is performed, the chemical etching method results in hydrogen termination, requiring carbon termination treatment (Patent Document 2). The ion implantation method and the sputtering method also require some kind of surface treatment because they are oxygen-terminated. In the liquid-phase synthesis method, surface termination treatment is usually performed after particle formation, so formation of any surface state is possible (Non-Patent Document 8). In this case, however, the size-controlled particles are produced from the bulk material prior to surface termination. In other words, the production of carbon-terminated silicon quantum dots, which are important for applications, usually requires a two-step process of particle production → surface treatment.

As a method for efficiently producing carbon-terminated quantum dots in contrast to the general production method described above, a laser irradiation method in an organic solvent proposed by Shirahata et al. is mentioned. In this method, a silicon single crystal substrate is irradiated with a high-energy-density pulse laser in an organic solvent (alkene such as 1-octene, alkyne) with unsaturated bonds. Laser irradiation causes silicon to evaporate and cool at the liquid/gas interface to produce nanoparticles. At this time, since the surface of the generated nanoparticles is in an active state, an addition reaction occurs with the unsaturated bond of the solvent, and the hydrocarbon group is terminated on the silicon surface. In this method, since particle generation and surface modification occur simultaneously, it can be said that this method is a simple one-step method as opposed to the usual two-step process. Moreover, the apparatus required is only a pulse laser, and expensive and large-scale apparatus are not required as compared with the ion implantation method and the sputtering method. Due to these advantages, a patent application (Patent Document 5) has also been filed using a similar technique.

However, the above-mentioned laser irradiation method produces nanoparticles with a very wide size distribution, making it difficult to control the size. In order to produce particles with different emission wavelengths, it is necessary to classify the dispersion solution of the produced particles. be. In addition, since a high-energy-density laser is required for particle generation, laser light is generally condensed using a condensing lens. Therefore, the irradiation area is several millimeters or less, and the problem is that the production amount (production yield) per process is very small.

In order to solve the above problems of producing carbon-terminated silicon quantum dots by the laser irradiation process, some of the present inventors and others proposed porous silicon, which is an aggregate of nanocrystalline silicon produced by a chemical etching method. was used as a target raw material for the in-liquid laser irradiation method (Patent Document 6, Non-Patent Document 10). It was demonstrated that this method increases the yield by about an order of magnitude compared to the conventional laser irradiation method using bulk silicon raw materials. This improvement in yield is believed to be due to the low thermal conductivity and low heat capacity resulting from the unique geometric structure of porous silicon. In addition, it was shown that the emission wavelength (size) of the generated quantum dots can be changed by controlling the nanostructure of the porous silicon material by changing the chemical etching conditions. Furthermore, they demonstrated that it is possible to increase size uniformity and yield by incorporating silicon-active materials, such as hydrogen fluoride, during the irradiation process.

JP-A-2009-91168 Japanese Patent Application Laid-Open No. 2010-188497 Japanese Patent Application Laid-Open No. 2013-095850 Japanese Patent Application Laid-Open No. 2009-911682 JP 2012-167175 A Japanese Patent Application Laid-Open No. 2016-216579

L. T. Canham Appl. Phys. Lett. 57, 1046, (1990) B. A. Kairdolf et al., Ann. Rev. Anal. Chem., 6, 143, (2013) J. R. Heath, Science, 258, 1131 (1992) T. Shimizu-Iwayama et al., J. Appl. Phys. 75, 7779 (1994) S. Charvet et al., J. Appl. Phys. 85, 4032 (1999) D. Kovalev, Phys. Status Sol. 215, 871 (1999) K. Kusova et al., 1 1300042 (2014) K. Dohnalova et al., J. Phys. Condens. Matter 26, 173201 (2014) Shirahata et al., Chem Commun. 4684 (2009) Nanotechnology 25, 275602 (2014), Appl. Phys. Lett. 108, 023105 (2016) J. Phys. Chem. C 121, 8623 (2017)

However, when viewed as an actual industrial process, the method previously invented by some of the inventors of the present invention is still insufficient in terms of environmental compatibility, cost reduction, high performance in light emission characteristics, and mass productivity. There is a need for further improvement of the fabrication method of environment-friendly silicon quantum dots fabricated by the in-liquid laser irradiation method. Specifically, in the method of fabricating silicon nanoparticles by laser irradiation using porous silicon as a raw material, (1) performance improvement (improvement of size controllability and luminous efficiency), (2) improvement of mass productivity ( (3) Cost reduction (reduction of laser irradiation raw material production cost).

The present invention uses an electrochemical method ( Porous silicon produced from a silicon wafer by the anodizing method) is used as a raw material. Porous silicon produced by anodization has significantly improved controllability of the porous nanostructure (porosity, silicon core size) compared to conventional porous silicon produced by pure chemical methods. Therefore, the present invention greatly improves the controllability of the emission (size) characteristics of nanocrystalline silicon quantum dots due to the high controllability of nanostructures in this electrochemical technique, and improves the performance of silicon quantum dots ( It is characterized by being able to improve size controllability and luminous efficiency). In addition, porous silicon produced by conventional pure chemical methods still contains a bulk layer within the porous silicon because the porous layer is formed by unstable chemical etching, and all the silicon raw materials are replaced by silicon quantum It was difficult to convert to dots, which caused a low production yield (ratio of produced colloid amount/raw material). On the other hand, since the porous silicon raw material produced by anodization in the present invention is composed only of a porous layer, it is considered possible to convert the raw material into silicon quantum dots with high yield. An improvement in yield can be expected, and this is a solution to the above problem (2). In addition, it is possible to prepare a large amount of porous silicon raw material in a single process by anodizing a large area wafer. In addition, the porous silicon raw material produced by the anodizing method is characterized by being composed of only the above-mentioned porous layer, which is even more outstanding than the porous silicon containing a bulk layer produced by a pure chemical method. Silicon quantum dots are expected to have low thermal conductivity and low heat capacity. Conversion is possible, and a drastic cost reduction can be expected due to the reduction in silicon raw material cost and production energy in the laser irradiation process of the above problem (3).

At present, silicon is a harmless and abundant material in the earth's crust. There are no commercial examples of silicon quantum dots using this, and if this is put on the market, it will have a great impact on the market. Since this technology can be produced by a simple production process of electrochemical etching and laser irradiation or heater heating, it has the effect of significantly reducing the price of the product. Currently commercially available compound-based semiconductor quantum dots (eg, CdTe, powder, particle size 1.5 nm, PlasmaChem)] are sold at about 1,000 yen/mg. On the other hand, the silicon quantum dots produced by the present invention use harmless silicon, which is abundantly contained in the earth's crust, and the raw material cost is low.

An example calculation of the expected manufacturing cost of silicon quantum dots according to the present invention is as follows.

(1) Porous silicon manufacturing process cost 1.0 yen/mg
[breakdown]
・Raw material (single crystal silicon wafer substrate) 500 yen/process (3.5g)
・Chemical etching chemicals (hydrofluoric acid, ethanol) 3,000 yen/process ・Electric chemical etching power ~4 yen/process (20 yen/kWh)
So, there is no problem with increasing the size of the equipment to the kg level.

(2)-1 Laser irradiation process Cost 8.1 yen/mg (Aiming for 1.0 yen in the future.)
[breakdown]
・Organic solvent 15 yen/process (100 mg for raw material)
・Consumables cost for pulse laser (flash lamp) 650 yen/process (100 mg for raw material)
・Silicon quantum dot production yield (quantum dot weight ÷ porous silicon weight) 0.85
・Pulse laser power supply power Up to 23 yen/process, upsizing equipment to the kg level is not a problem.

It is considered possible to further reduce the running cost by considering cost reduction by increasing the size of the laser irradiation device for mass production and by considering the use of semiconductor LD for the light source.

(2)-2Heater heating (300W) process cost ~0.4 yen [Breakdown]
・Organic solvent 15 yen/process (100 mg for raw material)
・Heater power supply cost ~20 yen/process (100 mg for raw materials)
・Silicon quantum dot production yield (quantum dot ÷ porous silicon weight) 0.85
So, there is no problem with increasing the size of the equipment to the kg level.

(3) Total silicon quantum dot production cost [in the case of laser irradiation process] ~10 yen/mg (aiming for around 2.0 yen/mg in the future)
[For heater heating irradiation process] ~2.0 yen/mg
As described above, it can be said that the quantum dot manufacturing method of the present invention has a manufacturing cost reduction effect of 1/100 to 1/1000 compared to commercially available compound semiconductor quantum dots at the technical development level. Furthermore, it has high biological and environmental safety, can be used in various fields, and has a high affinity with current silicon technology.

It is a figure which shows the production process flow of the porous silicon by this invention. (a) is a diagram showing emission spectra of various porous silicon flakes, and (b) is a diagram showing emission spectra of various silicon quantum dot colloidal dispersions. It is a figure which shows the production process flow of a silicon quantum dot colloid. FIG. 3 shows porous silicon flakes in organic solvent; FIG. 4 shows a UV laser irradiation process for porous silicon flakes in an organic solvent. FIG. 10 illustrates a heater heating process to porous silicon flakes in organic solvent. FIG. 4 is a diagram showing a silicon colloidal dispersion produced by a laser light irradiation process; FIG. 4 shows a silicon colloidal dispersion prepared by a heater heating process; FIG. 10 is a high-resolution transmission microscopy image showing that the colloidal particles produced are silicon single crystal quantum dots with a diameter of several nanometers. FIG. 4 is a diagram showing the results of analysis of surface bonding by Fourier transform infrared spectroscopy, showing that alkyl group terminations were formed. FIG. 10 shows a decrease in luminescence decay rate indicating that a good alkyl group termination has been formed.

As a specific example to which the present invention is applied, at present, silicon quantum dots with high emission quantum efficiency are manufactured by irradiating laser beams to porous silicon raw materials having various porosities prepared by electrochemical etching. did Details are as follows.

Porous silicon, which is the target raw material, is fabricated on the silicon wafer surface using a standard electrochemical etching method in a hydrofluoric acid aqueous solution (HF + H ₂ O + ethanol) as shown in the process flow shown in Figure 1. did. In this electrochemical etching, silicon atoms are locally eluted by anodic oxidation, and the etching reaction progresses from the surface to the inside one after another. As a result, nanometer-order micropores are formed on the surface and inside of the silicon wafer, and a porous silicon layer containing silicon quantum dots at high density is produced. After the etching is completed, the porous silicon is separated from the wafer for the laser irradiation process in liquid to obtain porous silicon pieces/powder (flakes) of millimeter to micrometer order.

In the electrochemical etching for producing the porous silicon flakes described above, the porosity (ratio of the space occupied by the total volume of the porous layer) and the silicon nanoparticles forming the porous layer can be easily controlled by the applied current density, the resistivity of the substrate, etc. The size of the crystal core can vary. As current examples, samples with porosity = ~70%, ~77%, ~80%, ~90% were prepared. Since these porous silicon wafers have silicon nanocrystal core layers of different sizes, they have different emission spectra and luminous efficiencies. FIG. 2(a) shows the measurement results of the emission spectrum of the porous silicon flakes. The broad emission band in the figure is due to electron-hole pair recombination due to quasi-direct transition from the indirect transition edge of the bandgap, which is shifted to high energy by the quantum size effect. From the figure, it can be seen that the emission spectrum peak position and peak width are different. Here, the porous silicon flake sample with a porosity of 70% is not plotted because it shows little luminescence. These differences in emission characteristics are considered to be due to changes in the size distribution of the silicon nanocrystal cores forming the porous layer of the porous silicon flakes, resulting in changes in the bandgap energy due to the quantum size effect. The sample with a porosity of 70% did not emit light because the core size was large and radiative recombination due to efficient quasi-direct transition due to the quantum confinement effect did not occur.

Subsequently, as shown in the process flow of FIG. 3, the porous silicon flakes obtained by the process of FIG. 1 are subjected to laser irradiation or heater heating by a hot plate in an organic solvent. Specifically, first, the porous silicon flakes are dispersed in 3 ml of 1-decene or 1-octene, which is an unsaturated hydrocarbon organic solvent, and 0.1 ml of HF is added (FIG. 4). This time, 10 mg of porous silicon flakes were used as the raw material. Then, as shown in FIGS. 5 and 6, while stirring with a magnetic stirrer, ultraviolet pulse laser light (wavelength: 266 nm, 5 ns pulse, 15 Hz) or heater heating with a hot plate (heater heating conditions 300 W, 35 °C). The energy of the pulsed laser light was about 1 mJ/pulse (0.5 J/cm ² ), and the heating temperature was 35°C. After irradiation, a small amount of precipitate (about 1 mg) is formed, which is removed with a centrifuge to obtain a yellow quantum dot colloidal dispersion solution (Figs. 7 and 8). When the weight of the colloid obtained here was measured, it was 9 mg. Therefore, it can be seen that the production yield (weight of quantum dots/weight of raw material) in this example is 85%. This value is remarkably improved compared to the yield (1% or less) of the production process in the invention (Patent Document 6) previously proposed by some of the present inventors, and is lower than the expected production yield. A significant improvement was confirmed.

Here, the reason for using an ultraviolet wavelength (266 nm) in the laser irradiation process is to terminate the alkyl group on the surface of the silicon quantum dots by a photochemical reaction with 1-octene. The Si—C bond energy is 3.83 eV (323 nm) (Non-Patent Document 8), and a laser beam with a wavelength of 323 nm or less is suitable for efficient termination on the Si surface and carbon molecules. Therefore, in the present embodiment, a pulsed laser of a 266 nm Nd:YAG laser with a frequency of 4 times was used as in the invention previously proposed by some of the present inventors (Patent Document 6). On the other hand, in the heater heating process, alkyl group surface termination (Non-Patent Document 8) is performed by a thermochemical reaction.

FIG. 2(b) shows measurement results of emission spectra of silicon quantum dot colloidal dispersions prepared from porous silicon flake raw materials (77 to 90%) with different porosities. From the figure, it can be seen that the emission spectrum changes by changing the raw material. Also, when 70% raw material was used, a colloidal dispersion exhibiting luminescence was not obtained. Here, the dependence of the emission peak wavelength on porosity shows a similar trend for both porous silicon flakes and Si particles. From the above, it is suggested that the size distribution of silicon quantum dots reflects the size distribution of the silicon core crystals of porous silicon, and that the quantum dots were generated by pulverization of the silicon core by laser irradiation or heater heating. It is It was also confirmed by high-resolution transmission microscope measurement that the colloidal particles produced were silicon single-crystal quantum dots with a diameter of several nanometers, as shown in FIG. As described above, in this example, it was shown that it is possible to easily control the emission color by the etching conditions of the porous silicon flakes, which is the starting material, which was expected.

The absolute quantum efficiency of light emission of silicon quantum dots fabricated from porous silicon flakes with a porosity of 80% was ~30%. This value is significantly improved compared to the quantum efficiency (~1%) of the starting material, porous silicon flakes, indicating that the alkyl group was terminated satisfactorily. The formation of this favorable alkyl group termination was also confirmed from the analysis of surface bonding with a Fourier transform infrared spectrophotometer (Fig. 10) and the decrease in the decay rate in the luminescence decay curve (Fig. 11).

Table 1 shows the results of comparison between the results of the above examples and the invention (Patent Document 6) previously proposed by some of the inventors of the present invention and the conventional laser irradiation process (Patent Document 2, Non-Patent Document 9). . As can be seen from the table, the present invention is a process that is remarkably superior in production yield and emission wavelength controllability compared to the usual laser irradiation process and the invention previously proposed by some of the present inventors (Patent Document 6). I know there is.

Claims (3)

A crystalline silicon wafer is immersed in a hydrofluoric acid aqueous solution, anodized using an electrochemical etching method, and etched from the surface toward the inside with HF contained in the hydrofluoric acid aqueous solution to form a nanometer-order porous layer. A porous silicon layer is produced, the porous silicon layer is separated from the silicon wafer in a liquid to obtain porous silicon flakes, and the obtained porous silicon flakes are used as a raw material in an organic solvent by laser irradiation, heating, or a combination thereof. and obtaining a quantum dot colloidal dispersion solution. In the electrochemical etching of porous silicon flake fabrication, the porosity or size of the silicon nanocrystalline core can be controlled by controlling the applied current density or the conductivity type and resistivity of the silicon wafer or the external light irradiation conditions including the dark state. 2. The method for producing silicon quantum dots according to claim 1, wherein the emission wavelength or emission efficiency is controlled by changing the emission wavelength. Using porous silicon flakes as a raw material, external energy is applied using LEDs in an organic solvent, such as a visible or infrared light source, a far infrared light source, a heat source, or a combination thereof. A method for producing the described silicon quantum dots.

Images