JP4504268B2 - Method for coating semiconductor surface and method for producing semiconductor particle - Google Patents

Description

  The present invention relates to a method for coating the surface of a group 4 semiconductor mainly containing a group 4 semiconductor element such as silicon, germanium, carbon, tin, lead, etc. with a group 4 element or a transition metal element, and nanoparticles using the method And a light emitting element using the same and an optical element such as a light modulation element based on an electro-optic effect.

    The research of silicon as an optical material aiming at application in the field of optical information communication is historically old and has been conducted for a long time from the mid-1980s to the present. However, compared to the research of newer optical materials such as III-V group compounds such as indium phosphide, gallium arsenide, and the insulator lithium niobate, there has been little progress in silicon research.

  There are two main reasons for this. The first reason is that silicon has no mechanism for emitting light, that is, it does not emit light. The second reason is that silicon does not show an electro-optic effect (EO effect) such as Pockels effect or Kerr effect. For these reasons, since the technical application range is limited, it is considered that silicon was not paid attention as an optical material. The EO effect is a general term for a phenomenon in which the refractive index changes when an external electric field is applied to a transparent material. When the refractive index change is proportional to the strength of the electric field, it is called the Pockels effect. When it is proportional to the square of, it is called the Kerr effect.

  By the way, the reason why silicon does not emit light is that it is a semiconductor material having an indirect band gap, and has a crystal structure that cannot make an efficient light emitting element such as a semiconductor laser. In contrast, indium phosphide and gallium arsenide, which are III-V compounds, are semiconductor materials having direct band gaps and are often used as semiconductor laser materials.

  The reason why silicon does not exhibit the EO effect is that the crystal structure has high symmetry. When a material exhibiting the EO effect is used, high-speed modulation of laser light becomes possible. More specifically, it becomes possible to make an optical modulator that converts an electrical signal into an optical signal. An optical modulator is actually made using lithium niobate exhibiting the Pockels effect, and is widely used in the field of optical fiber communication.

Table 1 summarizes the material technologies that have been studied in the past and give silicon a light emitting function. Six typical technologies are liquid phase synthetic silicon nanoparticles, rare earth ion doping, gas phase synthetic silicon clusters, and porous. Issues related to silicon, silicide semiconductors, and strained superlattices are shown.

  In conclusion, none of the material technologies has led to efficient light emission of silicon materials. As for the first liquid phase synthetic silicon nanoparticles, there is a problem of “small particle size problem” in which only particles having a small particle size can be produced. Since the silicon nanoparticles synthesized in the liquid phase have a particle size that is too small, they emit light mainly in the blue-violet region and it is difficult to obtain visible light emission.

  The second rare earth ion doping (Er ion doping) has a problem that light emitting Er ions are hardly dissolved in the silicon crystal, so that doping is difficult and as a result, it is difficult to obtain light emission.

  In contrast to the liquid phase synthetic silicon nanoparticles, the third gas phase synthetic silicon cluster has a problem of “large particle size problem” in which only particles having a large particle size can be produced. The cluster mainly emits light only in the infrared region and does not emit light in the visible region.

  The fourth porous silicon has a problem of “large particle size problem” as in the case of the vapor-phase synthetic silicon cluster. Since the particle size is still large in anodic oxidation, light emission in the red region is exhibited, but blue or green light emission is difficult to obtain.

The fifth silicide semiconductor (mainly β-FeSi ₂ ) has a problem that it does not shine unless strain is introduced into the material. Note that both calculations and experiments have shown that β-FeSi ₂ is an indirect bandgap semiconductor, and is essentially a non-luminous substance.

  The sixth strained superlattice (Si / SiGe) has a problem that particles having such a superlattice hardly emit light. SiGe is an indirect bandgap semiconductor in the first place, and therefore it is becoming difficult to emit light.

  As described above, the conventional material technology aiming to give a light emitting function to the silicon material cannot actually cause the silicon material to emit light efficiently.

  If silicon can be given a light emitting function, it will be possible to make significant progress in electronics for the reasons described below, and it is expected to contribute greatly to the realization of a prosperous future society.

  In the future society to come, the social desire to use information freely is expected to increase, and electronics such as LSIs are expected to have higher speed than ever. Information processing and information transmission players are expected to replace their current role of electrons with the light with the fastest information speed.

  In the field of optical information communication, when individual optical devices (semiconductor lasers, optical modulators, etc.) that support optical fiber communication in terms of hardware are replaced with silicon, LSI and optical fiber are integrated on a silicon wafer. The time to assimilate will come soon. This means that electronics and communication are integrated, and the speed of information processing and transmission can be increased to the limit of the speed of light. Therefore, siliconization of optical devices will greatly increase the speed of various applications and social infrastructures such as personal computers, search (Internet), image recognition systems, communications, forecasting / prediction (calculations and calculations), etc., which require a wide bandwidth. It is a translation.

  However, as described above, the conventional material technology has an essential problem that silicon cannot be efficiently illuminated.

  The present invention has been made on the basis of recognition of such a problem, and the object of the present invention is to provide a novel semiconductor surface coating that is completely different from the conventional method as a method for imparting a light emitting function to a group 4 semiconductor material. The present invention provides a method and a method for producing a novel liquid phase synthetic nanoparticle based on the same technique, and provides an optical element based on a Group 4 semiconductor material having a light emitting function.

The method for coating a semiconductor surface according to the present invention comprises:
On the surface of the semiconductor material containing a group 4 element selected from the group consisting of silicon, germanium, carbon, and tin, a reducing agent soluble in an inert organic solvent and having an electron donating property is contained in the inert organic solvent. The surface of the semiconductor material is coated with a metal derived from the reducing agent, and then reacted with an electron withdrawing group-containing compound containing an element selected from the group consisting of a group 4 element and a transition metal element. The surface is coated with a group 4 element or a transition metal element derived from the electron-withdrawing group-containing compound.

  The method for producing semiconductor particles according to the present invention is characterized in that a seed crystal is used as a semiconductor material and crystal growth is performed by the above-described method.

  The semiconductor particles according to the present invention are manufactured by the above-described method for manufacturing semiconductor particles.

  The light-emitting device according to the present invention is a light-emitting device including a light-emitting portion that generates light by electrical or optical energy excitation, and the light-emitting portion includes the semiconductor particles. Is.

  The light modulation element according to the present invention is a light modulation element including a light modulation unit that generates an optical signal by electrical or optical light modulation, and the light modulation unit includes the semiconductor particles. It is characterized by this.

  According to the method of the present invention, it is possible to form a coating whose thickness and composition are precisely controlled at the atomic level on the semiconductor surface. Furthermore, by applying the method to seed particles, synthesis of Group 4 semiconductor nanoparticles with controlled particle size and controlled composition becomes possible for the first time. Compared to Group 4 semiconductor nanoparticles produced by other manufacturing methods, Group 4 semiconductor nanoparticles with excellent emission characteristics such as tailored emission wavelength and improved emission efficiency can be realized for the first time. .

  And when the group 4 semiconductor nanoparticles produced by the production method of the present invention are used, the optical device can be siliconized, and the speed of LSI advances, so a personal computer, a search (Internet), an image recognition system, Various applications and social infrastructure such as communication, forecasting and prediction (calculation and calculation) can be realized at very high speed, and the industrial merit is great.

Hereinafter, embodiments (embodiments) of the present invention will be described in detail with reference to the drawings.
First of all, in Section 1, in order to clarify the advantages and differentiation points of the present invention, a solution of nanoparticles called the Grignard method, which is one of the conventional methods for imparting a light emitting function to a Group 4 semiconductor material, is described. The principle of the phase synthesis method will be described by taking the synthesis of silicon nanoparticles as an example. Next, Section 2 examines the “small particle size problem”, which is a drawback of the Grignard method, and explains that it is difficult to grow to a particle size capable of visible light emission. Next, in Sections 3-a and 3-b, the principle of the nanoparticle production method of the present invention is explained, and it is possible to precisely control the particle size at the atomic level. It is shown that it is possible to realize visible light emission with Group 4 semiconductor nanoparticles, which is one of the objects of the present invention. In section 3-c, the principle of the semiconductor surface coating method of the present invention is explained, and it is stated that the light emitting function can be imparted to the target group 4 semiconductor material.

1) Grignard method This method is a technique for obtaining nanoparticles by three-dimensionally growing Si-Si bonds using a condensation reaction between silicon compound molecules in a solution.

FIG. 1 schematically shows the formation and growth of Si—Si bonds. As shown in the figure, the starting materials used in the Grignard method are silicon chloride (liquid) and magnesium metal (powder). As the inert organic solvent, dehydrated ethylene glycol dimethyl ether (hereinafter referred to as “glyme”) is generally used. The reaction steps in glyme are as follows.
1. Silicon chloride is added into glyme containing metal magnesium powder.
2. By stirring, silicon chloride and metal magnesium collide.
3. By this collision, magnesium chloride molecules, which are reaction intermediates, are generated in situ.
4). Cl ₃ Si—MgCl + Cl ₄ Si → Cl ₃ Si—SiCl ₃ + MgCl ₂ that forms a Si—Si bond by molecular condensation reaction of silicon chloride and magnesium chloride represented by the following reaction formula
5). While continuing stirring, Si—Si bonds are grown three-dimensionally.

  As can be seen from the reaction formula of the reaction step 4, the Si—Si bond is the reaction point of the silicon chloride molecule —the MgCl group of the magnesium chloride silicon molecule generated in situ in the glyme— It is formed by a condensation reaction with a Cl group. In this chemical reaction, the —MgCl group is an electron-donating functional group, and the —Cl group is an electron-withdrawing functional group.

  Note that magnesium silicon chloride molecules are classified into type 1 to type 4 depending on the number of added magnesium atoms. The generation probability by type strongly depends on the concentration of silicon chloride molecule and magnesium metal, but under normal conditions (where the molar ratio of silicon chloride molecule: metal magnesium is 1: 1 to 1: 4), 1 is generated at a higher rate than the other types 2, 3, and 4. This can be explained as follows. That is, the magnesium chloride silicon molecules are generated when the silicon chloride molecules dissolved in the solution collide with metallic magnesium powder that is insoluble in the solution. The probability that the generated magnesium silicon silicon molecules will collide with metallic magnesium again is very low, and it reacts with other silicon chloride molecules dissolved in the solution before that and with the growing nanoparticles. For this reason, type 1 magnesium chloride silicon molecules are produced at a higher rate than other types.

  Each type 1 magnesium chloride silicon molecule has one reaction point (≡Si-MgCl) and three reaction points (≡Si-Cl), and the number of reaction points and non-reaction points is one. I have not done it. As will be described below, this discrepancy causes a small particle size problem.

2) The small particle size problem of the Grignard method FIG. 2a schematically shows the atomic structure of crystalline silicon. Crystalline silicon has a diamond structure and has a six-membered ring as shown in the figure.
FIG. 2b schematically shows silicon nanoparticles in the early stage of growth. This figure shows a state in which Si—Si bonds are grown from the first generation (first) silicon atom, which is the nucleus of the growth reaction, to the third generation silicon atom.

  Similar to crystalline silicon, silicon nanoparticles grow in a diamond structure, so that the magnesium chloride silicon molecules that should become fourth-generation silicon atoms react with two third-generation silicon atoms to form a six-membered ring. Need to form. For this reason, the magnesium chloride silicon molecule that reacts with the third generation silicon nanoparticles needs to react with the reaction sites (≡Si—Cl) of two third generation silicon.

  However, the type 1 magnesium chloride silicon molecule usually produced has only one magnesium atom and has only one reactive site (≡Si—MgCl). Accordingly, it can be seen that the magnesium chloride silicon molecule has a unique situation in which it cannot reliably react with either one of the third generation silicon atoms.

  Now, suppose that the fourth generation silicon has grown by reacting with one of the third generation silicon. However, the steric hindrance between the three —Cl groups added to the other unreacted third generation silicon atom and the three —Cl groups added to the fourth generation silicon atom is very large. The Si-Si bond between the third and fourth generations is energetically unstable, and this bond tends to be fragile. For this reason, under normal synthesis conditions (conditions where the molar ratio of silicon chloride molecule: metallic magnesium is 1: 1 to 1: 4), the fourth generation silicon is extremely difficult to grow.

  If nanoparticle growth stops until the third generation, the nanoparticle will consist of only 17 atoms. When the particle size of the nanoparticles at this time is estimated, it is approximately 0.5 nm to 1 nm. When the peak wavelength of photoluminescence is examined, it is about 350 to 400 nm, which corresponds to the size of particles emitting in the ultraviolet-blue-violet region.

  According to the literature, in order to obtain blue, green, and red light emission by making silicon into nanoparticles, it is necessary to adjust the particle size to 2 to 2.5 nm, the particle size 2.5 to 3 nm, and the particle size 3 to 5 nm, respectively. It is estimated. The larger the emission wavelength, that is, the larger the emission color changes in the order of ultraviolet → blue → green → red, is due to the quantum confinement effect.

  From the above points, it can be seen that the particle size obtained by the Grignard method is too small to obtain visible light emission. This is a small particle size problem peculiar to the Grignard method. Similar suppression of nanoparticle growth reaction is also observed in other group 4 elements such as germanium, carbon, and tin, and is a phenomenon common to the synthesis of group 4 semiconductor nanoparticles using the Grignard method.

3) The method for producing semiconductor nanoparticles according to the present invention and the method for coating the semiconductor surface Unlike group 4 semiconductor nanoparticles, metal nanoparticles such as Au, Ag, Pt, Pt, Cu, Fe, Co, Ni, CdS For II-VI group compound semiconductor nanoparticles such as CdSe, CdTe, ZnS, ZnSe, ZnTe, and III-V group compound semiconductor nanoparticles such as GaAs, InAs, and InP, a liquid phase synthesis method has been established. This is because constituent atoms such as Au, Ag, Pt, Pt, Cu, Fe, Co, Ni, Cd, Zn, S, Se, Te, Ga, In, As, and P are in solution. This is because they can exist relatively stably in an atomic ion state, and crystal growth to metal nanoparticles or compound semiconductor nanoparticles can be controlled relatively easily by basic experimental parameters such as time, reagent concentration, and temperature.

  On the other hand, group 4 atoms such as Si, Ge, C, and Sn are extremely unstable in atomic ion state except for a plasma state created by CVD or the like. It exists only in the form of a compound molecule. For this reason, the chemical reaction in the liquid phase synthesis needs to be assembled based on a molecular condensation reaction, unlike nanoparticles of other materials.

  The essence of Si-Si bond formation in the previous Grignard method is that one electron-donating reactive site (≡Si-MgCl) in one molecule and one electron-withdrawing reactive site in another molecule ( It was in a molecular condensation reaction with (≡Si—Cl). The problem of small particle size, which is a problem, is due to the fact that the number of electron-donating groups and electron-withdrawing groups are inconsistent in magnesium chloride silicon molecules and silicon chloride molecules involved in Si-Si bond formation.

  In order to realize a liquid phase synthesis method for freely synthesizing Group 4 semiconductor nanoparticles, in other words, a new liquid phase synthesis method capable of crystal growth to a desired particle size with a desired chemical composition and light emission at a desired wavelength. It is necessary to redesign the entire chemical reaction from the ground up so that the number of electron-donating groups and electron-withdrawing groups of the molecule to be subjected to the condensation reaction always match.

  Therefore, the method for coating a semiconductor surface according to the present invention comprises a surface of a semiconductor material containing a group 4 element selected from the group consisting of silicon, germanium, carbon, and tin, is soluble in an inert organic solvent, and is an electron. A donor reducing agent is reacted in the inert organic solvent to coat the surface of the semiconductor material with a metal derived from the reducing agent, and then contains an element selected from the group consisting of group 4 elements and transition metal elements The electron withdrawing group-containing compound is reacted, and the surface of the semiconductor material is coated with a group 4 element or a transition metal element derived from the electron withdrawing group-containing compound.

  FIG. 3 is a schematic explanatory view of the most basic method for producing a group 4 semiconductor nanoparticle in a liquid phase, in which the method for coating a semiconductor surface according to the present invention is applied to a group 4 semiconductor particle. The horizontal axis represents the procedure for introducing the reagent (synthesis procedure), and the vertical axis represents the particle size of the nanoparticles corresponding to this procedure.

  In the method according to the present invention, a compound molecule containing a group 4 atom which is an electron withdrawing group-containing compound (such as silicon chloride) and an electron-donating reducing agent which is soluble in an inert organic solvent (such as sodium naphthalene) The charge transfer complex or a rare earth iodide such as samarium diiodide) is added to the solution alternately and repeatedly as necessary to coat the semiconductor surface. If this method is applied to seed particles made of a semiconductor, group 4 semiconductor nanoparticles can be produced. The method according to the present invention makes it possible for the first time to produce Group 4 semiconductor nanoparticles whose particle size is precisely controlled and whose composition is controlled at the atomic level. The group 4 nanoparticles produced in this manner can emit light with a high emission efficiency at a desired emission wavelength as compared to conventional nanoparticles.

  In the following, the principles of 3-a) a quaternary group 4 semiconductor nanoparticle manufacturing method composed of a single element and 3-b) a method of manufacturing a binary quaternary compound semiconductor nanoparticle will be described. . As an example, the former deals with silicon nanoparticles and the latter with germanium / carbon compound nanoparticles.

3-a) Method for Producing Group 4 Semiconductor Nanoparticles Containing One- Element System One method for producing semiconductor particles according to the present invention is one in a group 4 semiconductor material selected from silicon, germanium, carbon, and tin. group 4 atom a and four with a molecule consisting of an electron withdrawing group X of the (electron-withdrawing group-containing compound) and AX ₄ is a MY one alkali atoms M and a charge transfer complex consisting of one aromatic molecule Y of Then, when chemically reacting the molecule and the charge transfer complex in an inert organic solvent, each of them is repeatedly added in the order of AX ₄ and MY or in the order of MY and AX ₄ ; Particles, particularly semiconductor nanoparticles, can be grown by controlling the size of a three-dimensional nanostructure composed of A atoms in atomic units.

As an example, a case where silicon nanoparticles are formed will be described. According to the synthesis procedure shown in FIG. 3, in glyme, silicon chloride as a silicon atom source and sodium naphthalene charge transfer complex as a reducing agent for reducing silicon chloride (exactly (sodium) ⁺ (naphthalene) ^-. a charge transfer complex or less, by adding alternately and glyme solution of NN as complex) can be 1 atomic generation by one crystal growth of silicon nanoparticles having a diamond structure.

More specifically, it is as follows.
An NN complex is gradually added into glyme in which silicon chloride is dissolved in advance, and a silicon nanoparticle of a generation is generated when it first becomes a seed crystal. The first nanoparticle that becomes the seed crystal is a nanoparticle whose surface is terminated with chlorine. When the NN complex is continuously added, the NN complex further acts on the surface of the nanoparticle, the surface chlorine atoms are reduced to the form of NaCl and eliminated, and the naphthalene molecule of the reducing agent is oxidized from a negative ion to a neutral ion. As a result, it dissolves into the solution, and finally the surface of the nanoparticles is entirely covered with sodium atoms derived from the NN complex. When the surface of the sodium-terminated nanoparticle is made to react with silicon chloride molecules, the surface sodium atoms are oxidized to form NaCl and desorbed, while silicon chloride loses chlorine atoms and is reduced, resulting from silicon chloride. The silicon atoms are chemically bonded to the surface of the nanoparticle, and a silicon atom layer in which the silicon atom layer is grown for exactly one atom generation is obtained. In other words, a monoatomic film made of silicon atoms is formed on the surface of the particle. Thereafter, if NN complex and silicon chloride are added N times in a regular order, silicon nanoparticles can be grown by N atom generations, that is, the grain size can be selectively increased by N atom generations. Can do.

  In addition to the nanoparticles, NaCl and neutralized naphthalene molecules are mixed in the solution, but both of them are chemically changed to a stable state, and thus no longer react with the nanoparticles.

FIG. 4 shows a calculation result of the particle diameter D with respect to the generation G of diamond structure silicon nanoparticles prepared by the manufacturing method according to the present invention. The particle diameter D (nm) is a converted diameter when the nanoparticles are regarded as a sphere, and can be estimated by the following equation.
D = a × [3M / 4π] ^1/3 (1)
a is the lattice constant (0.543 nm) of bulk crystalline silicon, and M is the total number of silicon atoms constituting the nanoparticle. Table 2 shows the M values of diamond-structured nanoparticles.

  As can be seen from the figure, when one nanoparticle is grown by this method, the particle size can be increased by about 0.3 nm. Visible emission of silicon nanoparticles is generally considered to occur at 10 nm or less, particularly between 2 and 5 nm in particle size. Therefore, it is understood that the nanoparticles should be grown in the 7th to 16th generations. The generation number of the first seed crystal can be adjusted according to the synthesis conditions (temperature, concentration, stirring speed, etc.). It is easy to obtain seed crystals of 2 to 6 generations, and if the synthesis conditions are high temperature, low concentration and high speed stirring, it is possible to adjust to seed crystals of a small generation and uniform size. It is also possible to use larger nanoparticles produced by the method of the present invention, for example, 7 to 16th generation nanoparticles as seed crystals.

The use of a reducing agent that is soluble in an inert organic solvent and is an electron-donating reducing agent such as sodium naphthalene or samarium diiodide is an important factor for realizing the production method according to the present invention. One. In the above description, the NN complex, which is a charge transfer complex, is used as a reducing agent. However, an electron-donating reducing agent such as a rare earth iodide can also be used, which is soluble in other inert organic solvents. . Table 3 shows a comparison table of various reducing agents.

  There are two types of reducing agents: an electron donating reducing agent that gives electrons to the partner and a proton donating reduction that gives protons. With proton-donating reducing agents, it is possible to replace the chlorine atoms covering the nanoparticle surface with hydrogen atoms, but silicon chloride and hydrogen-terminated silicon nanoparticles do not react in a low temperature solution, Care must be taken because growth may not be controlled.

  Table 3 also shows that there are types of reducing agents that are insoluble and soluble in the solvent. With a reducing agent that is insoluble in the solvent, the probability that the nanoparticles and the reducing agent collide is almost zero, and the nanoparticle surface remains chlorine-terminated. Chlorine-terminated silicon nanoparticles are difficult to react with silicon chloride, so it is still difficult to control nanoparticle growth.

  In order to solve this problem, in the present invention, a reducing agent that is soluble in an inert organic solvent and is electron-donating is used. As such a reducing agent, a charge transfer complex (a combination of an electron-donating metal atom and an electron-accepting organic molecule) or a rare earth iodide is preferable. More specifically, sodium-naphthalene complex, lithium-naphthalene complex, lithium-di-t-butylbiphenyl, etc. as charge transfer complexes, samarium diiodide, yttrium diiodide, diiodide as rare earth iodides Examples include europium. The rare earth iodide is particularly preferably a divalent rare earth iodide. Here, “soluble in an inert organic solvent” means that an amount sufficient to carry out the method of the present invention is dissolved in the inert organic solvent. Therefore, since the required solubility varies depending on the type of the inert organic solvent used in the method of the present invention and the type of the reducing agent, the specific solubility is not limited. However, if an example of the solubility of the reducing agent used for this invention is given, it is preferable that the solubility with respect to glyme is 0.0005 mol / liter or more, and it is more preferable that it is 0.001 mol / liter or more.

  In the NN complex, like magnesium, sodium itself is usually a metal powder that is insoluble in an inert organic solvent. However, by combining the complex with a naphthalene molecule that is soluble in the solvent, the sodium remains in the solvent while maintaining its reducing ability. It can be solubilized.

  Other charge transfer complexes other than the NN complex can be used as a reducing agent in the method of the present invention as long as they are electron-donating and soluble in a solvent. In addition to sodium, the electron-donating metal atom can be selected from alkali metal atoms such as lithium, potassium, rubidium and cesium. As the electron-accepting organic molecule to be combined, in addition to naphthalene, pyrene, Various aromatic organic molecules such as anthracene, tetracene, perylene, phenanthrene, fluorene, biphenyl and the like can be selected from organic molecules containing in their molecular skeleton. Lithium naphthalene complexes are one promising reducing agent.

Moreover, rare earth iodide is mentioned as a reducing agent which can be used for the method of this invention. Here, the rare earth iodide is a reducing agent that utilizes the valence change of the rare earth metal ions. The reaction formula when such a reducing agent is used is as follows when samarium diiodide (3.2 equivalents) is used as the halogenated silane with respect to silane bromide (1 equivalent).
SiBr ₄ + 3.2 * SmI ₂
→ Si nanoparticles (halogen terminated) + 3.2 * SmI _3-x Br _x (2)

  Since rare earth iodide is a reducing agent that utilizes valence change, unlike charge-transfer complexes such as NN complexes, it does not generate aromatic molecules as a by-product, so it is easy to extract and purify after processes. There is.

  As a supply source of silicon atoms, in addition to silicon chloride, silicon bromide, silicon iodide, tetramethoxysilane, tetraethoxysilane, or the like can be selected.

  As the inert organic solvent, an inert organic solvent that does not react with the above-described reducing agent or silicon chloride is essential. As such a solvent, a solvent having an ether bond such as glyme, tetrahydrofuran, or diethyl ether can be used.

  In addition to the silicon nanoparticles described here, germanium nanoparticles, diamond nanoparticles, tin nanoparticles, and the like can be basically produced by the same procedure. Germanium nanoparticles can be synthesized from germanium chloride and NN complexes, diamond nanoparticles can be synthesized from carbon tetrabromide and NN complexes, and tin nanoparticles can be synthesized from tin chloride and NN complexes. In any case, it is possible to selectively grow crystals one generation at a time.

3-b) Method for Producing Group 4 Compound Semiconductor Nanoparticles Consisting of Binary System The difference from the method for producing semiconductor particles comprising a single element is that the source of each group 4 atom constituting the group 4 compound This is the point of using two different molecules.

That is, another method for producing semiconductor particles according to the present invention provides a group 4 semiconductor material selected from silicon, germanium, carbon, tin, and lead, one group 4 atom A and four groups, and an electron withdrawing group X. AX ₄ is a molecule composed of a group 4 and B ₄ is a molecule composed of a group 4 atom B different from A and the electron withdrawing group X, and charge transfer is composed of one alkali atom M and one aromatic molecule Y. When the complex is MY, when the molecule and the charge transfer complex are chemically reacted in an inert organic solvent, each of them is represented by AX ₄ , MY, BX ₄ , MY or MY, BX ₄ , MY. , AX ₄ or BX ₄ , MY, AX ₄ , MY, or MY, AX ₄ , MY, BX ₄ In atomic units The size can be controlled and the particles can be grown.

Hereinafter, germanium / carbon compound nanoparticles will be described as an example.
FIG. 5 is a schematic diagram illustrating this binary system synthesis method. Similar to FIG. 3, the horizontal axis represents the synthesis procedure and the vertical axis represents the generation of atoms. As can be seen from the figure, germanium tetrachloride → NN complex → carbon tetrabromide → NN complex is one cycle, and these reagents are added regularly and repeatedly, so that germanium atom generation and carbon atom generation are alternately arranged. It becomes possible to grow germanium-carbon compound nanoparticles having a diamond structure (cubic zinc sulfide structure).

  More specifically, it is as follows. In FIG. 5, since the synthesis is started from germanium tetrachloride, the seed crystal becomes germanium nanoparticles. As in the previous one-dimensional case, the seed crystal surface is initially chlorine-terminated, but changes to sodium-terminated by continuing to add the NN complex. Next, when carbon tetrabromide is added, surface sodium atoms are desorbed in the form of NaBr, while carbon tetrabromide loses bromine atoms and is reduced and chemically bonded to the surface of the nanoparticle, just one atomic generation worth of carbon. Nanoparticles with the outermost surface terminated with chlorine, with atoms covering germanium seed crystals, are obtained. Thereafter, if NN complex, germanium tetrachloride, and carbon tetrabromide are continuously added in a regular order, germanium-carbon nanoparticles in which germanium atoms and carbon atoms are grown alternately are obtained.

  The first seed crystal germanium nanoparticles can be adjusted to a small, uniform seed crystal within several generations by adjusting the synthesis conditions (temperature, concentration, stirring speed, etc.). .

  By using the method according to the present invention, it is possible to produce nanoparticles of regularly arranged group 4 compounds composed of two or more types of group 4 elements except for the seed crystal portion. In general, the synthesis of such regularly arranged compounds is extremely difficult. This is because, for example, in a germanium-carbon compound, both germanium and carbon are tetracoordinate atoms, and are mixed well with each other, so that it is difficult to take a regular arrangement, and an irregular arrangement is likely to occur.

  In addition to germanium / carbon compound nanoparticles, binary materials such as silicon / germanium compounds, silicon / carbon compounds, silicon / tin compounds, silicon / lead compounds, germanium / tin compounds, germanium / lead compounds, tin / lead compounds, etc. It is possible to create nanoparticles of group 4 compounds.

  Furthermore, in addition to one or more kinds of Group 4 elements, transition metal elements other than Group 4 elements, preferably titanium, vanadium, zirconium, hafnium, manganese, and the like, which can be tetracoordinate elements, can be selected. According to such a method, silicon-titanium compound, silicon-vanadium compound, silicon-zirconium compound, silicon-hafnium compound, silicon-manganese compound, germanium-titanium compound, germanium-vanadium compound, germanium-zirconium compound, germanium- Hafnium compounds, germanium / manganese compounds, carbon / titanium compounds, carbon / vanadium compounds, carbon / zirconium compounds, carbon / hafnium compounds, carbon / manganese compounds, tin / titanium compounds, tin / vanadium compounds, tin / zirconium compounds, tin / zirconium compounds Hafnium compounds, tin / manganese compounds, lead / titanium compounds, lead / vanadium compounds, lead / zirconium compounds, lead / hafnium compounds, lead / manganese compounds, etc. Nano-particles can be created.

  Depending on the application, binary nanoparticles having an arrangement different from the regular arrangement can be prepared. Assuming that one atom of the binary system is an A atom and the other is a B atom, in addition to the regular arrangement... ABAB..., An arbitrary atomic arrangement for each generation, for example, AABBAABB, or ABABAAB , Or a completely irregular array... AABABBBBBABAABBBB.

  In a ternary system or a multi-element system, it is also possible to create a group 4 nano-particle that is regularly or irregularly arranged.

That is, another method for producing semiconductor particles according to the present invention is based on a group 4 semiconductor material selected from silicon, germanium, carbon, tin, and lead. comprising a molecule with AX _4, wherein a and different types of group 4 atoms B and a molecule consisting of an electron withdrawing group X and BX _4, a molecule consisting of group 4 atoms C and the electron-withdrawing group X of a different type and the AB CX ₄ , a molecule composed of a group 4 atom D and an electron withdrawing group X different from ABC is DX ₄ , a molecule composed of a group 4 atom E and an electron withdrawing group X different from ABCD is EX _4, and 1 When the charge transfer complex consisting of one alkali atom M and one aromatic molecule Y is MY, when the molecule and the charge transfer complex are chemically reacted in an inert organic solvent, the molecule AX ₄ , BX ₄ , CX ₄ , DX ₄ , EX ₄ and two or more kinds of molecules are selected, and each of them is added two or more kinds by repeatedly adding the molecules, the order of the charge transfer complexes, or the charge transfer complex, the order of the molecules. It is possible to grow a particle by controlling the size of a three-dimensional nanostructure composed of any group 4 atom of each atomic unit.

In addition, another method for producing semiconductor particles according to the present invention includes a group 4 semiconductor material selected from silicon, germanium, carbon, tin, and lead, from one group 4 atom A and 4 and an electron withdrawing group X. A charge transfer complex comprising AX ₄ as a molecule, ZX _{4 as a} molecule composed of one transition metal atom Z and four electron-withdrawing groups X, and one alkali atom M and one aromatic molecule Y when was a MY, in an inert organic solvent, when for chemically reacting the charge transfer complex with the molecule, _{respectively,} AX 4, MY, _ZX 4, MY of ranking or _MY,, ZX 4, MY, A three-dimensional nanostructure consisting of two types of AZ atoms can be added by repeatedly adding the AX ₄ sequence, or ZX ₄ , MY, AX ₄ , MY sequence, or MY, AX ₄ , MY, ZX ₄ sequence. Control the size in atomic units, grain Child can grow up.

  By such a method, silicon, germanium, carbon nanoparticles and the like can be prepared.

  Furthermore, it is also possible to replace part of the one-component, two-component, and multi-component nanoparticles with luminescent impurity atoms. For example, lanthanoid rare earth ions, iron group transition metal ions, and the like. Since rare earth ions are usually tricoordinate, they cannot bond to one of the four adjacent silicon atoms. Therefore, it is possible to compensate for this by adding a pentacoordinate atom such as a phosphorus atom.

  From the above, by the method of the present invention, the synthesis of Group 4 semiconductor nanoparticles whose particle size is controlled to 2 to 5 nm in terms of a sphere, which is precisely obtained at the atomic level, preferably in terms of sphere, and whose composition is controlled can be synthesized. It becomes possible for the first time. Compared with Group 4 nanoparticles according to the prior art, Group 4 nanoparticles excellent in emission characteristics such as tailor-made emission wavelength can be realized for the first time.

3-c) Growth Method on Silicon Wafer According to the method of the present invention, not only can Group 4 semiconductor nanoparticles be synthesized in solution, but also various Group 4 semiconductors can be formed into films or particles on a silicon wafer. It becomes possible to grow crystals. The procedure is as follows.

  First, a silicon wafer is introduced into a synthesis chamber, and the surface is cleaned in an ultrahigh vacuum state to deposit silicon atoms.

  Next, break the vacuum, return to atmospheric pressure, and inject glyme into the synthesis chamber. At this time, in order to avoid surface contamination, a sufficiently dry inert gas having a sufficiently low dew point is introduced.

Next, the wafer surface is brought into contact with a halogen such as bromine (Br ₂ ) or an acid such as odorous acid (HBr) or hydrochloric acid (HCl) to terminate the silicon atoms on the surface with halogen. In general, since the surface of an untreated wafer is not terminated with halogen, sodium atoms are easily adsorbed by bonding halogen to the group 4 element on the wafer surface by such treatment.

The method for halogenating the wafer surface can also be realized by a method different from the above. For example, for chlorine termination, after treatment with hydrofluoric acid (HF) or ammonium fluoride (NH ₄ F) and once hydrogen termination of the wafer surface, benzoyl peroxide or the like is used as a radical initiator, Treatment in chlorobenzene with the appropriate amount of phosphorus pentachloride (PCl ₅ ) or chlorine (Cl ₂ ) results in a chlorine terminated surface. As for bromine termination, hydrogen termination is performed once, and then a bromotrichloromethane (CBrCl ₃ ) treatment is performed using benzoyl peroxide or the like as a radical initiator to obtain a bromine terminated surface.

  Next, an NN complex is added to adsorb sodium atoms on the surface silicon atoms. In short, the silicon wafer surface is terminated with sodium.

  Next, in the case of growing a single-system group 4 semiconductor such as silicon on a wafer, silicon chloride and an NN complex may be added repeatedly in accordance with the procedure up to the previous section. Further, when growing an ordered binary group 4 semiconductor, for example, a germanium / carbon compound, in accordance with the procedure up to the previous section, a cycle of germanium tetrachloride → NN complex → carbon tetrabromide → NN complex is used as one cycle. Reagents may be added regularly and repeatedly.

  In addition, it is also possible to replace a part of an irregularly arranged binary compound, a ternary or more multi-component compound, and a uni-component / bin-ary / multi-component compound with luminescent impurity atoms. .

  The shape of the finally obtained composite strongly depends on the areal density of sodium atoms to be adsorbed first. That is, film growth occurs when sodium atoms are densely adsorbed on the silicon surface, and particle growth occurs when they are sparse. Therefore, it is important to control the amount of adsorption of sodium atoms. This can be controlled by monitoring the adsorption of sodium atoms in real time by infrared reflection spectroscopy or the like.

Synthesis of Silicon Nanoparticles FIG. 6 schematically shows a synthesis apparatus used in the present invention. The four-necked flask 1 of the reaction vessel has openings 2 and 3, and the inside 5 is purged with argon gas (dew point −80 degrees or less) and sufficiently dried. A drip funnel or a cooling pipe (not shown) is attached to the opening 2 or 3. In the production method of the present invention including this example, atmospheric components such as water and oxygen are directly linked to the contamination of the nanoparticles, and therefore it is necessary to perform synthesis in a flask in which such an atmosphere is controlled.

  In order to synthesize silicon nanoparticles, 100 ml of an anhydrous glyme solution in which 5 mmol of silicon chloride is dissolved is prepared in a flask. While maintaining this solution at room temperature and stirring with a stirrer 6 at a speed of 600 revolutions per minute, a 0.2 mol anhydrous glyme solution of NN complex is slowly dropped from the opening 2 using a dropping funnel.

  Although the anhydrous glyme solution of the NN complex is a dark green liquid, when the NN complex solution is dropped into the colorless and transparent silicon chloride glyme solution, the solution 4 in the flask first changes to a cream color. Further, the NN complex was continuously added, and when the dripping amount of the NN complex reached approximately 20 mmol, the solution suddenly changed to dark red.

  When the solution in the flask begins to turn deep red and the color change reaches a steady state, silicon chloride is then slowly added dropwise. When the drop volume reaches approximately 1 mmol, the solution turns cream again.

  Thereafter, using the solution color in the flask as an indicator, when the color changes to cream, the addition of silicon chloride is stopped and the NN complex solution is added. When the color changes to dark red, the addition of NN complex solution is stopped and silicon chloride is added again. repeat. By removing the silicon chloride previously charged in the flask and repeating this operation four times in the order of dropping the NN complex solution and dropping the silicon chloride, it is possible to synthesize chlorine-terminated silicon nanoparticles.

  In the cream-colored solution, chlorine-terminated silicon nanoparticles, NaCl, and neutralized naphthalene molecules are mixed. In the chlorine-terminated nanoparticle, electrons are attracted from the silicon atom to the chlorine atom, and it is assumed that the silicon atom is partially positively charged.

  On the other hand, it is considered that sodium-terminated silicon nanoparticles are generated in the deep red solution. Negatively charged silicon anions are known to exhibit a red body color. From this, in sodium-terminated silicon nanoparticles, it is presumed that electrons are supplied from sodium atoms to silicon atoms, and the silicon atoms are partially negatively charged and exhibit a red body color. The addition of the NN complex turns the solution dark red, which is indirect evidence that the nanoparticle surface is sodium terminated.

  When the chlorine-terminated silicon nanoparticles generated in the flask are taken out into the atmosphere, hydrolysis easily occurs and the silicon nanoparticles are oxidized. For this reason, the surface stabilization treatment is subsequently performed in the flask.

  Using a dropping funnel, 200 mmol of butyl alcohol is slowly dropped into the flask and stirred at room temperature for 24 hours. By doing so, silicon nanoparticles covered with butoxy groups and surface-stabilized can be produced in the flask. As by-products, NaCl and naphthalene are generated.

  100 ml of hexane is poured into the flask and stirred sufficiently to selectively extract silicon nanoparticles and naphthalene. The extracted hexane solution can be washed three times with pure water using a separatory funnel to remove and purify the remaining NaCl. The purified hexane solution can be dehydrated by immersing it in magnesium sulfate.

  Thereafter, naphthalene and hexane are respectively removed and purified by vacuum distillation to obtain the target silicon nanoparticles.

  By observation with an electron microscope, it is possible to confirm that the average particle size of the silicon portion of the silicon nanoparticles synthesized in this way is approximately 2.1 nm. According to the compositional analysis of the constituent elements, both chlorine and oxygen show a low value of about the detection limit of the device in weight ratio (wt%), so it can be said that this is a preparation method for obtaining highly pure nanoparticles. it can. Compared with nanoparticles made by the Grignard method, the amount of each impurity atom (Cl is 1 wt%, O is 10 wt%) can be greatly reduced.

  The obtained silicon nanoparticles can be dissolved in hexane, and the fluorescence characteristics can be examined. From the emission spectrum obtained by ultraviolet light excitation, a peak emission wavelength of 525 nm and a spectrum half width of about 60 nm are obtained. Luminescent quantum efficiency can be determined based on rhodamine 6G dye. According to the measurement, it is possible to realize silicon nanoparticles having very excellent light emission characteristics, in which efficiency is a value of about 30% to 60%.

Particle size control In the second example, in order to control the particle size of the silicon nanoparticles, the number of drops of each of the NN complex solution and silicon chloride is increased from 4 times to 12 times. In the same procedure, silicon nanoparticles are synthesized.

  By observation with an electron microscope, it can be confirmed that the silicon nanoparticles synthesized in this way have an average particle size of the silicon portion of about 3 nm. From the fluorescence spectrum obtained by photoexcitation, a spectrum having a peak emission wavelength of 630 nm and a spectrum half width of about 80 nm is measured. It is possible to realize silicon nanoparticles having a large particle diameter and having very excellent light emission characteristics, in which the light emission quantum efficiency is about 10% to 40%. According to the present invention, the particle size can be controlled by the number of times of dropping, and the emission peak wavelength can be controlled.

EL Light-Emitting Element FIG. 7 is a cross-sectional view schematically showing a light-emitting element of an example of the present invention. This light-emitting element can be formed on a silicon wafer. The p ⁻ Si layer 8 provided on the p ⁺ Si layer 7 as the p electrode, the amorphous Si layer 12 as the opposing n electrode, and the silicon nanoparticle as the active layer 9. Each can be composed of a packed bed of particles. The light emitting device of FIG. 7 further includes a SiO ₂ layer 10 and a pad electrode 11. The light emitting element is electrically connected to an electrode extraction wiring (not shown).

The particle size of the silicon nanoparticles used for the active layer can be selected according to the emission wavelength. Here, blue light-emitting nanoparticles having an average particle size of 2.1 nm are cited as an example.
This light-emitting element can be driven at 20 mA and 4 V, and it is possible to obtain a very good value of 5 cd / A as the EL luminous efficiency.

Wafer coating In this embodiment, a germanium-carbon compound is coated on a silicon wafer. The synthesizer used in the present invention has the same configuration as that shown in FIG. 6 in principle. The difference is that the reaction vessels are all made of metal in order to support ultra-high vacuum, and a heater is used to adjust the wafer temperature. Is added.

  Prior to the synthesis, the flask in the reaction vessel on which the 1 cm square wafer is placed is evacuated to an ultrahigh vacuum, and the wafer temperature is raised in this atmosphere to perform a surface cleaning process to remove impurity atoms other than silicon atoms. The cleaning process conforms to a normal LSI process. Next, the vacuum in the flask is broken to return to atmospheric pressure, and 100 ml of anhydrous glyme is injected. At this time, purging with sufficiently dried argon gas (dew point minus 80 degrees or less) to avoid surface contamination.

In order to terminate the silicon surface with sodium, a 0.1 molar anhydrous glyme solution of NN complex is slowly dropped, and the solution is stirred at 600 revolutions per minute to cause the NN complex to collide with the wafer surface and react. Since there are approximately 10 ¹⁵ silicon atoms (about 2 nanomoles) on the surface of a 1 cm square silicon wafer, an excessive amount of 1 millimole of NN complex is dropped in order to terminate sodium densely.

  In order to coat this sodium-terminated wafer surface with a germanium-carbon film, these reagents are regularly and repeatedly added in one cycle of germanium tetrachloride → NN complex → carbon tetrabromide → NN complex. In addition, an excessive amount of any reagent is dropped so that the surface reaction proceeds sufficiently. Since carbon tetrabromide is a solid reagent, it is dissolved in glyme in advance and a 0.1 molar solution is prepared.

  After repeating this dropping cycle 100 times, in order to stabilize the surface of the synthesized germanium / carbon film, 1 mol of butyl alcohol is slowly dropped and stirred at room temperature for 24 hours. In this way, a germanium carbon film covered with butoxy groups and stabilized, which is the target, can be formed on the wafer surface.

  By removing the wafer from the flask and washing with pure water and hexane, naphthalene, NaCl, unreacted reagents and other by-products adhering to the wafer can be removed. In this way, a wafer coated with a germanium-carbon film, which is the object, can be obtained.

  By observing the cross section with an electron microscope, it can be confirmed that the film thickness of the synthesized coating layer is between about 20 nm and 30 nm. According to the compositional analysis of the constituent elements, germanium and carbon are respectively detected from the coating layer, and the ratio is shown to be Ge: C = 1: 1. In addition, chlorine, oxygen, and sodium have values as low as the detection limit of the apparatus, and thus can be said to be a production method that can produce a highly pure film.

  Incidentally, when the obtained film was excited with ultraviolet rays, an emission spectrum having a peak in the near infrared region was obtained. The emission quantum efficiency is about 50%, and a Group 4 compound semiconductor having excellent emission characteristics can be realized.

  In the fifth embodiment, a procedure similar to that of the first embodiment is performed except that the synthesis is performed using a reducing agent different from the NN complex, lithium di-t-butylbiphenyl charge transfer complex (LDBB complex). Synthesize silicon nanoparticles.

  By observation with an electron microscope, it is possible to confirm that the silicon nanoparticles synthesized in this way have an average particle size of the silicon portion of 2.2 nm. From the fluorescence spectrum obtained by photoexcitation, a peak emission wavelength of 530 nm and a spectrum half width of about 60 nm are obtained. Luminescence quantum efficiency shows a value of about 30% to 50%, and it is possible to realize silicon nanoparticles having a large particle diameter and having very excellent emission characteristics. According to this example, it can be seen that, besides the NN complex, a charge transfer complex composed of an alkali metal and an aromatic organic molecule can be used as the reducing agent.

  In the sixth example, silicon nanoparticles are synthesized by changing the types of the reducing agent and the silane compound in the same procedure as in the first example. Silicon particles are synthesized using samarium diiodide as the reducing agent and silicon bromide as the silane compound.

  After stabilizing the surface of the silicon nanoparticles in the same procedure as in the first example, the extracted and purified nanoparticles are observed with an electron microscope. As in the first example, the average particle size is 2.0 nm. Can be confirmed. The fluorescence characteristics are also the same as in the first embodiment.

  This example shows that it is possible to use rare earth iodide as a reducing agent instead of the charge transfer complex.

The conceptual diagram explaining the conventional manufacturing method (Grignard method). The schematic diagram explaining the small particle size problem of the conventional manufacturing method (Grignard method). (A) Structure of bulk crystalline Si (b) Structure of Si nanoparticles The schematic diagram explaining the synthesis | combination of 1 type | system | group group 4 semiconductor nanoparticle in connection with the manufacturing method of this invention. The graph which shows the calculation result of the particle size with respect to the generation of the silicon nanoparticle produced by this invention. The schematic diagram explaining the synthesis | combination of the binary system 4 group compound semiconductor nanoparticle in connection with the manufacturing method of this invention. 1 is a cross-sectional view schematically showing a synthesis device according to a first embodiment of the present invention. Sectional drawing which represents roughly the light emitting element concerning the 3rd Example of this invention.

Explanation of symbols

1 Four-necked flask 2, 3 Opening 4 Reaction solution 5 Inside of flask (dry argon)
6 Stirrer 7 p ⁺ Si layer 8 p ^- Si layer (p electrode)
9 Active layer (silicon nanoparticle packed layer)
10 SiO ₂ layer 11 Pad electrode 12 Amorphous Si (n electrode)

Claims (8)

The surface of the semiconductor material comprising silicon, the electron donating reductant dissolved in an inert organic solvent beat drops, the intermediate having a surface coated with metal derived from the reducing agent of a semiconductor material An electron withdrawing group-containing compound containing an element selected from the group consisting of silicon, germanium and carbon is dropped onto the intermediate, and the surface of the semiconductor material is a group 4 element or transition derived from the electron withdrawing group-containing compound. A method for coating a semiconductor surface, comprising coating with a monoatomic film of a metal element.   The semiconductor surface coating method according to claim 1, wherein the reducing agent is a charge transfer complex or a rare earth iodide.   The reducing agent is selected from the group consisting of sodium naphthalene complex, lithium naphthalene complex, lithium di-t-butylbiphenyl, samarium diiodide, yttrium diiodide, and europium diiodide. Or the method for coating a semiconductor surface according to 2.   The semiconductor surface according to any one of claims 1 to 3, wherein the reducing agent and the electron-withdrawing group-containing compound are repeatedly reacted with each other to coat with a group 4 element or a transition metal element a plurality of times. Coating method.   The semiconductor surface coating method according to claim 1, wherein the electron-withdrawing group-containing compound contains the same element as the element contained in the semiconductor material.   The semiconductor surface coating method according to claim 4, wherein a part of the plurality of coatings is coated with an electron-withdrawing group-containing compound containing an element different from the electron-withdrawing group-containing compound used for the other coating.   The method for coating a semiconductor surface according to any one of claims 1 to 6, wherein the group 4 element atom on the surface is terminated with a halogen prior to the reaction of the reducing agent.   A method for producing semiconductor particles, wherein a crystal seed growth is performed by the method according to any one of claims 1 to 7, using a particulate seed crystal as a semiconductor material.

Images