Technical field

This disclosure relates generally to heavy-metal-free metal chalcogenide nanoplatelets and methods of synthesising the same.

Background

Atomically thin quasi-two-dimensional (2D) colloidal semiconductor

nanoplatelets are of significant importance. Despite intense interest in these materials, the syntheses of heavy-metal-free colloidal semiconductor nanoplatelets are still greatly undeveloped and the mechanism behind their highly anisotropic shape and precise atomic thickness remains unclear.

Being analogous to graphene and transition metal dichalcogenides, atomically thick 2D semiconductor nanocrystals are emerging as a type of intriguing materials and of broad utility due to their properties such as exceptionally narrow photoluminescence, low laser threshold, small Stokes shift, and ultrafast exciton dynamics. Unfortunately, despite intense efforts in the past 10 years, current 2D semiconductor nanocrystals still face several challenging issues. Firstly, the growth mechanisms of nanoplatelets with a wurtzite structure are still unclear although there are a few studies on the growth mechanism of zinc blended nanoplatelets. Secondly, most of the reported 2D semiconductor nanocrystals contain highly toxic cadmium and lead ions. Also, the synthesis of nanoplatelets utilizes toxic organic phosphines as the ligands/solvents and low reaction yields of a few tens of to a few hundreds of milligrams are generally obtained. Lastly, the synthesis requires tedious multistep reactions, such as preparation and purification of the seeds, hot injection of precursors at an elevated temperature, rigorous experimental conditions such as oxygen- and moisture-free environment, and suffers from batch-to-batch variability. All of the above issues impede the large-scale commercialization of nanoplatelets.

It is to be understood that, if any prior art publication is referred to herein, such reference does not constitute an admission that the publication forms a part of the common general knowledge in the art, in Australia or any other country.

Summary

Disclosed is a method of forming a metal chalcogenide nanoplatelet, comprising: providing elemental chalcogenide, a metal salt and a solvent; converting the elemental chalcogenide to a chalcogen anion; mixing the chalcogen anion with the metal salt in the solvent; and heating the chalcogen anion and the metal salt to a temperature that is sufficient for the chalcogen anions to react with a metal ion of the metal salt to form the metal chalcogenide nanoplatelet.

The elemental chalcogenide and metal salt may be mixed together before converting the elemental chalcogenide to the chalcogen anion so that the chalcogen anion is formed in the presence of the metal salt.

Disclosed is a method of forming a metal chalcogenide nanoplatelet, comprising: providing elemental chalcogenide, a metal salt and a solvent; converting the elemental chalcogenide to a chalcogen anion in the presence of the metal salt and the solvent; and heating the chalcogen anion and metal salt to a temperature that is sufficient for the chalcogen anions to react with a metal ion of the metal salt to form the metal chalcogenide nanoplatelet.

The solvent may react with the elemental chalcogenide to form the chalcogen anion at a temperature below the temperature that is sufficient for the chalcogen anion to react with the metal ion to form a metal chalcogenide nanoplatelet. The temperature sufficient for the chalcogen anion to react with the metal ion may range from about 130 °C to about 250 °C, such as about 130 °C to about 170 °C, or less than about 200 °C. The method may further comprise using a reducing agent to convert the chalcogenide to the chalcogenide anion.

The method may be performed in the presence of oxygen.

The metal ion may exclude heavy metal ions so that the metal chalcogenide nanoplatelet is substantially heavy metal free. The metal chalcogenide may be free from at least one of Cd, Pb and Hg.

In an embodiment, at the temperature sufficient for the chalcogen anion to react with the metal ion to form a metal chalcogenide nanoplatelet, the chalcogen anion may react with the metal ion to firstly form bundled nanowires, which fuse to form fragmented nanobelts. The nanobelts may then be converted to small nanosheets and then to the final metal chalcogenide nanoplatelets.

The solvent may be an organic solvent. The solvent may be anhydrous. An embodiment may further comprise heating the solvent to a temperature sufficient to drive water out of the solvent. The temperature sufficient to drive water out of the solvent may be about 110 °C. The solvent may be heated at the temperature sufficient to drive water out of the solvent for about 1 minute to about 60 minutes. The solvent may include an alkylamine and/or arylamine.

The nanobelt may have a thickness of about 0.99 nm, and the nanosheet and nanoplatelet both may have a thickness of less than about 50 nm, such as 1.40 nm, for example 1.39 nm.

The method may be a one-pot wet-chemical method.

An embodiment may further comprise extracting any unreacted chalcogenide and/or chalcogen once the metal chalcogenide nanoplatelets have formed. The chalcogenide and/or chalcogen may be extracted using an organophosphine.

An embodiment may further comprise diluting the solvent with a second solvent after the metal chalcogenide nanoplatelets have formed. The second solvent may be a halogenated solvent.

An embodiment may further comprise isolating the metal chalcogenide nanoplatelets. The metal chalcogenide nanoplatelets may be isolated from the mixture by precipitation. Precipitation may be performed by diluting the mixture with a solvent in which the metal chalcogenide nanoplatelet is insoluble in. The solvent in which the metal chalcogenide nanoplatelet may be insoluble in may solubilise any unreacted elemental or ionic chalcogenide. The solvent may be a polar solvent including methanol.

An embodiment may further comprise cooling the solvent prior to isolating the nanoplatelets.

The metal salt may include Ag ⁺ , Zn ²⁺ , Mo ^x+ , Cu ^x+ , Bi ^x+ and/or Sb ^x+ salts. The metal salt may further include Hg ^x+ .ln an embodiment, the Zn ²⁺ salts may include Zn(NC>3)2, ZnC , Zn(CH3COO) ₂ and Znl _å , the Mo ^x+ salts may include M0CI5, the Bi ^x+ salts include BiCI _3, and the Sb ^x+ salts may include SbCI ₃ . The elemental chalcogenide may include powders of Se, S and/or Te. The metal chalcogenide nanoplatelet may have a formula including ZnS, ZnSe, ZnTe, ZnS _x Sei- _x , ZnSe _x Tei- _x , Sb S ₃ , CU S and Bi ₂ S ₃ . More than one chalcogen anion may be reacted with the metal salt so that an alloyed metal nanoplatelet is formed. The nanoplatelet may have a formula of MS _Q Se _T Tei- _(Q+T) , wherein M is a metal of the metal salt and Q and T range from 0 to 1

An embodiment may further comprise subjecting the metal chalcogenide nanoplatelet to cation exchange by reacting the metal chalcogenide nanoplatelet with a second metal salt at a temperature sufficient to allow exchange of cations in the metal chalcogenide nanoplatelet. The temperature sufficient to allow exchange of cations in the metal chalcogenide nanoplatelet may range from about 25 °C to about 200 °C, such as about 100 °C to about 200 °C. The second metal salt may include Cd ^x+ , Mo ^x+ and Cu ^x+ .

An embodiment may further comprise: forming a mixture comprising a metal nanoparticle precursor and the metal chalcogenide nanoplatelet; and heating the mixture to convert the nanoparticle precursor to a nanoparticle. The nanoparticle may be formed on a surface of the metal chalcogenide nanoplatelet. Forming the mixture may comprise forming a suspension of the metal nanoparticle precursor and the metal chalcogenide nanoplatelet in a solvent. The solvent may be a non-polar solvent, such as a short chain alkane including hexane. The solvent may include a reducing agent. The reducing agent may be an amine-containing solvent. More than one reducing agents may be used.

Also disclosed is a nanoplatelet formed using a method as set forth above.

Also disclosed is a metal chalcogenide nanoplatelet, comprising a metal and one or more chalcogenides, the metal chalcogenide having a formula of MX _Q YTZI- _(Q+T) , wherein M is a metal, X, Y and Z are chalcogenides, and Q and T range from 0 to 1 ; and wherein the metal chalcogenide has a thickness less than about 50 nm, such as less than about 10 nm. The formula may include MC _n Z ₍ i- _n) , wherein X and Z are chalcogenides and y ranges from 0 to 1 .

The metal chalcogenide may be free from at least one of Cd, Pb and Hg.

The thickness of the metal chalcogenide may be uniform and may have a size less than about 5 nm including a thickness of about 1 .40 nm. The thicknesses may be about 1 .39 nm. Lateral dimensions of the metal chalcogenide may range from about 5 to 100 times, such as from about 15 to 40 times, the thickness of the nanoplatelet. The lateral dimensions of the metal chalcogenide may range from about 30 nm to about 80 nm.

The metal chalcogenide may have a wurtzite crystal structure. The metal chalcogenide may comprise two or more metals. In an embodiment, M may be Zn, Cu, Mo, Bi and/or Sb. The metal M may further include Hg. In an embodiment, X and Z may include S, Se and/or Te. In an embodiment, the metal chalcogenide may have a formula including ZnS, ZnSe, ZnTe, ZnS _x Sei_ _x , ZnSe _x Tei_ _x , Cu ₂ S, MoS ₂ , Sb ₂ S ₃ , Bi ₂ S ₃ and MSoSe _T Tei-t _Q+T) . The nanoplatelet may be substantially rectangular in shape. The nanoplatelet may further comprise nanoparticles arranged on an outer surface of the nanoplatelet. The nanoparticles may be formed from Fe, Co, Ni, Au, Pt, Pd, Ru, Rh and/or Ir. Also disclosed is an electronic device comprising the nanoplatelet as set forth above. The electronic device may include a light detector such as a UV and/or light detector.

Also disclosed is a method of forming a photodetector comprising depositing the nanoplatelets as set forth above onto an electrode. The step of depositing the nanoplatelets onto the electrode may comprise depositing a solution containing the nanoplatelets onto the electrode and subsequently evaporating the solution. The method may further comprise annealing the photodetector after depositing the nanoplatelets onto the electrode. The method may further comprise surface treating the electrode prior to deposition of the nanoplatelets.

Also disclosed is a photodetector formed using the method as set forth above.

Detailed description of embodiments

Disclosed is a method of forming a metal chalcogenide nanoplatelet. The method comprises providing elemental chalcogenide, a metal salt and a solvent. The solvent is generally an organic solvent. A mixture of the elemental chalcogenide, metal salt and solvent can be considered to be a reaction mixture. The elemental chalcogenide is converted to chalcogen anions. For example, if the chalcogenide is sulphur, it would be converted to the sulphide ion. Once the chalcogen ion is formed, it is then mixed with the metal salt in the solvent, i.e. the reaction mixture is mixed. The chalcogen anion and the metal salt are heated so that the chalcogen anion reacts with metal ion of the metal salt to form the metal chalcogenide. The terms metal chalcogenide, metal chalcogenide nanoplatelet(s) and nanoplatelet(s) are used interchangeably.

The elemental chalcogenide can be mixed with the metal salt before the elemental chalcogenide is converted to the chalcogen anion. This can simplify the method since multiple steps and vessels can be eliminated. In other embodiments, the chalcogen anion is formed first and then it is reacted with the metal ion of the metal salt. For example, when the chalcogenide is tellurium, the tellurium is converted to the tellurium anion e.g. with a hydride source and then mixed with the metal salt.

Also disclosed is a method of forming a metal chalcogenide nanoplatelet. This method comprises providing elemental chalcogenide, a metal salt and a solvent. The elemental chalcogenide is then converted to the respective chalcogen anion in the presence of the metal salt and the solvent. Once the chalcogen anion is formed, both it and the metal salt are heated to a temperature sufficient for the chalcogen anion to react with the metal ion to form a metal chalcogenide nanoplatelet.

In either method, the specific temperature required for the chalcogen anion and metal ion to react will depend on the type of chalcogen anion and metal salt, and whether or not an alloyed metal chalcogenide nanoplatelet is formed. For example, the temperature at which the chalcogen anion and metal ion react together may range from about 130 °C to about 200 °C, such as 140 °C to about 170 °C. Temperatures above 200 °C tend to promote phase transition of the nanoplatelet into nanowires, although this upper temperature limit does depend on the specific chalcogenide and metal salt so some embodiment may be heated to a temperature higher or lower than 200 °C. To form the nanoplatelet, the chalcogen anion first reacts with the metal ion to form bundled nanowires, which then convert into fragmented nanobelt with a thickness of about 0.99 nm. The nanobelt is then converted into a nanosheet, which is then converted into the metal chalcogenide nanoplatelet. In some embodiments both the nanosheet and nanoplatelet have a thickness of about 1.39 nm. The energy (i.e.

temperatures) required to increase the thickness of the nanoplatelet causes a phase transition into other morphologies. This means that for embodiments where both the nanosheet and the nanoplatelet has a thickness of about 1.39 nm, the nanoplatelets are generally limited to a thickness of about 1.39 nm using the method. Other embodiments have a thickness less than about 50, such as less than about 10 nm, such as less than about 5 nm. Such specific and uniform thickness may be desirable in some embodiments since nanoplatelets used in electrical applications generally require a narrow size distribution to ensure that the properties of the nanoplatelet do not vary. As the band gap (or the emitting wavelength) of nanoplatelets is mainly depended on their thickness, thickness control of nanoplatelets may help to tune the colour of emitted light from the nanoplatelets. Further, since the method may only produce nanoplatelet having a uniform thickness, for example less than 5 nm such as about 1.39 nm, this means that there may be little thickness variation between batches. The resulting nanoplatelet may also have a generally rectangular morphology. That is, on an X-Y plane, one dimension is larger than the other. This is compared with e.g. square and hexagonal nanoplatelets which extend evenly in the X-Y direction.

The chalcogen anion may form at a temperature below the temperature that is sufficient for the chalcogen anion to react with the metal ion to form a metal chalcogenide nanoplatelet. Given the chalcogen anion and metal ion may begin to react with one another at temperatures greater than 130 °C, the temperature at which the elemental chalcogenide is converted to the chalcogen anion may be less than 130 °C. However, in some embodiments, the temperature at which the elemental chalcogenide is converted to the chalcogen anion may be higher than the temperature that is sufficient for the chalcogen anion to react with the metal ion to form a metal chalcogenide nanoplatelet. In these embodiments, the rate limiting step will tend to be the conversion of element chalcogenide to the chalcogen anion.

The solvent may react with the elemental chalcogenide to form the chalcogen anion. The solvent may act as a catalyst or the solvent may be consumed during the conversion. Since the elemental chalcogenide is reduced to form the chalcogen anion, the solvent may act as a reducing agent. Put another way, the solvent may act as an electron donor. The solvent may be an organic solvent. In this way, the method can be considered to be a colloidal chemical synthetic method, which differs from nanoplatelet prepared using aqueous solution methods. The method may further comprise using a reducing agent to convert the chalcogenide to the chalcogenide anion. For example, such a step may be required when the relative reactivity of the solvent and

chalcogenide is not sufficient to convert the chalcogenide to the chalcogenide anion. When the chalcogenide includes Te, a reducing agent may be required in addition to the solvent to form Te anions. The reducing agent may include a phosphine-based reducing agent and/or super hydride. Generally, with aqueous solution methods, the resulting morphologies of the nanoplatelet have thicknesses in the order of greater than 10 nm, and lateral dimensions of a few hundred nm. This is different to the morphologies of the nanoplatelets prepared by the disclosed methods, having thickness of about 5 nm, such as about 1.4 nm, and lateral dimensions of about 30 nm to about 80 nm. A biphasic solvent system may be used. For example, an oil-in-water or water-in-oil solvent system may be used. A biphasic system may utilise a hydrothermal method of producing nanoplatelets.

The solvent may include an alkylamine and/or arylamine. The alkylamine may be a short chain (i.e. C6-C12 or greater) alkylamine such as hexylamine, octylamine, decylamine, dodecylamine, hexadecylamineand oleylamine. As an example, when an alkylamine is used as the solvent, elemental sulphur-alkylamine solutions exist mainly as alkylammonium polysulfides. Upon heating to the temperatures required for nanocrystal growth, the polysulfide ions react with excess alkylamine to liberate H2S, which combines with the metal ion of the metal salt to form metal chalcogenide nanoplatelets. A similar mechanism is used for selenium.

Since long chain alkylamines have a non-polar body with a relatively polar head, they may also act as a surfactant. Use of a surfactant can help to control the morphologies of the nanoplatelets. The presence of surfactants generally helps to reduce the size of the dimensions of the morphologies compared to, for example, aqueous solution methods that do not use a surfactant. The surfactants can help to prevent aggregation of the nanoplatelet during their formation, which can help to produce nanoplatelet with dimensions having a narrow size distribution. The surfactants may also help to ensure that the size of the nanoplatelet is within set parameters, for example below a thickness limit. The surfactant may cover a surface of the nanoplatelet once formed and isolated from the reaction mixture.

Moisture can affect the formation of the nanoplatelet. To overcome this, the solvent may be anhydrous. However, when non-anhydrous solvents are used, the method may include an additional step of heating the solvent to a temperature sufficient to drive water out of the solvent. Since water boils at 100 °C (at normal atmospheric conditions), heating the solvent above 100 °C, for example at 110 °C, may help to drive water out of the solvent. However, the temperature required to drive water out of the solvent is partially determined by the type of solvent. For relatively non-polar solvents, temperatures just above 100°C may be sufficient to drive off the water out of the solvent. However, for more non-polar solvents such as dimethylformamide and dimethylsulfoxide temperatures in excess of 100 °C may be needed, such as 150 °C. The solvent can be heated above the boiling point of water for a time long enough to drive out of the water from the solvent. Depending on how“wet” the solvent is initially, the required time may be short, for example a few minutes, to an hour or two. In some embodiments, the solvent is heated at the temperature sufficient to drive water out of the solvent for about 1 minute to about 60 minutes, such as about 10 minutes to about 30 minutes. Being able to simply drive off water out of the solvent rather than requiring anhydrous solvents may be advantageous since it can help to reduce the costs and complexities of the method of making the metal chalcogenide nanoplatelet. This advantage may be even more so for large scale synthesis.

The method may be performed in the presence of oxygen. Not requiring inert conditions may help to simplify the method since the expense and expertise required to perform reactions under inert conditions is not needed. Compared to traditional synthesis of nanoplatelets which require inert and anhydrous conditions, being able to carry out the method in the presence of oxygen may be a significant advantage, especially at larger scales where multi-gram quantities are produced.

The method may be carried out in a single reaction vessel, i.e. the method may be a one-pot wet-chemical method where all reagents are mixed in the single vessel. Such a reaction set-up can simplify the method and allow commercial quantities of nanoplatelets to be synthesised. In an embodiment the method is performed in a continuous flow reactor. Traditional nanoplatelet synthesis requires the use of specialised equipment and multi-vessel reaction systems. Further, traditional nanoplatelet synthesis is limited to tens to low hundred milligrams quantities. Because the method may be carried out simply in a one-pot step in the presence of oxygen and using non-anhydrous solvents as a starting reagent, the method is robust and simple and in some embodiments provides multi-gram quantities, such as greater than 10 grams of nanoplatelet. Compared with traditional metal chalcogenide nanoplatelet synthesis, this is a significant improvement as it provides a method to allow nanoplatelets to be synthesised at commercial scales.

Generally, the chalcogenide is added in excess so that the metal salt is the limiting reagent, which means that there is generally some unreacted chalcogenide and/or chalcogen anion remaining in the reaction mixture once the nanoplatelets are formed. An embodiment may therefore further comprise extracting any unreacted chalcogenide and/or chalcogen once the metal chalcogenide nanoplatelets have formed. Unreacted chalcogenide and/or chalcogen anion may be removed by extraction in a suitable solvent. To assist in the removal of unreacted chalcogenide and/or chalcogen anion, any unreacted chalcogenide may be converted to the chalcogen anion. In an embodiment, the chalcogenide and/or chalcogen may be extracted using an organophosphine. The organophosphine may be trioctylphosphine (TOP), trimethylphosphine, triethylphosphine, tributylphosphine (TBP),

trihexylphosphine, tricyclohexylphosphine, triphenylphosphine, tritolylphosphine, dimethylbutylphosphine, dimethylhexylphosphine, dimethyloctylphosphine, dibutylmethylphosphine, dibutylhexylphosphine, dibutyloctylphosphine,

dihexylmethylphosphine, dihexylbutylphosphine, dihexyloctylphosphine,

dioctylbutylphosphine, dioctylhexylphosphine, ditolylphenylphosphine,

diphenyltolylphosphine and/or mixtures thereof. An embodiment uses trioctylphosphine (TOP) and/or tributylphosphine. As an example, when the chalcogenide is selenium, the phosphine may be used to remove any unreacted selenium powder from the reaction mixture. The type of organophosphine used may be determined by its reactivity relative the chalcogenide and/or nanoplatelet.

Once the nanoplatelets are formed, they may be isolated from the reaction mixture. Therefore, an embodiment may further comprise isolating the metal chalcogenide nanoplatelets. Isolation may be achieved by extraction and/or precipitation. To assist in removing any impurities and to help precipitation, the solvent (i.e. the reaction mixture) may be diluted with a second solvent after the metal chalcogenide nanoplatelets have formed. The second solvent should generally be soluble in the solvent and reaction mixture, but the second solvent may in some embodiments be insoluble in the solvent of the reaction mixture. Removal of impurities and/or precipitation of the nanoplatelets may rely on partitioning of any unreacted starting materials between the two insoluble solvents. In one embodiment, the second solvent may be a halogenated solvent such as dichloromethane, chloroform, carbon tetrachloride.

The metal chalcogenide nanoplatelets may be isolated from the mixture by precipitation, generally by diluting the reaction mixture with a solvent in which the metal chalcogenide nanoplatelet is insoluble in. The reaction mixture may be diluted by adding the reaction mixture to the solvent in which the metal chalcogenide nanoplatelet is insoluble in. Alternatively, the reaction mixture may be diluted with the solvent in which the metal chalcogenide nanoplatelet is insoluble in. To help purify the

nanoplatelets, the solvent in which the metal chalcogenide nanoplatelet is insoluble in may solubilise any unreacted elemental or ionic chalcogenide. For example, any unreacted chalcogenide and/or chalcogen anion may be soluble in the solvent in which the metal chalcogenide nanoplatelet is insoluble in. For example, methanol may be used to solubilise any unreacted sulphur whilst at the same time precipitate metal sulphide nanoplatelets. The solvent in which the metal chalcogenide nanoplatelet is insoluble in may be a polar solvent such as methanol, ethanol, and propanol including its isomeric forms and/or butanol and its isomeric forms. A mixture of two or more solvents may be used to precipitate the nanoplatelets. An embodiment used methanol to precipitate the nanoplatelets. To assist in isolating the nanoplatelets, the solution containing the formed nanoplatelets (i.e. the reaction mixture or the reaction mixture dissolved in the second solvent) may be cooled prior to isolating the nanoplatelets. For example, the reaction mixture may be diluted with chloroform, then cooled, then diluted with methanol to precipitate the nanoplatelets.

The metal ion may exclude heavy metal ions so that the metal chalcogenide nanoplatelet is substantially heavy metal free. The majority of research on 2D nanoplatelets is mainly limited to CdSe primarily due to its facile synthetic accessibility. However, as cadmium is highly toxic, cadmium based system poses a substantial threat to human health and the environment. Eliminating the need to use heavy metals in the synthesis of the metal chalcogenides nanoplatelet makes the method more environmentally benign compared to traditional methods. It may also help to reduce the costs of the method as wastes containing heavy metals do not need to be disposed of. The heavy metals may include Cd, Pb and Hg. However, heavy metal-based nanoplatelets can have beneficial properties for electronic application, so in some embodiments the nanoplatelets include heavy metals such as Cd, Pb and/or Hg.

The metal salt may include Zn ²⁺ , including ZnO, Zn(N0 ₃ ) ₂ , ZnCI ₂ , Zn(CH ₃ COO) ₂ , Znl ₂ , ZnBr ₂ , zinc stearate, zinc acetylacetonate, zinc undecylanate, zinc 2- ethylhexanoate, diethylzinc, zinc oleate, zinc phosphonic compounds, zinc fluoride, dimetylzinc, and zinc carbonate. Mo ^x+ may include MoCI ₅ , bis(acetylacetonate) dioxomolybdenum, molybdenum isopropoxide, and bis(ethylbenzene) molybdenum, MO(CO) ₆ , and cycloheptatriene molybdenum tricarbonyl. Cu ^x+ may include CuCI, CU(N0 ₃ ) ₂ , CUS0 ₄ , CU(CH ₃ COO) ₂ and copper acetyloacetate. Bi ^x+ may include BiCI ₃ , BiBr ₃ , Bil ₃ , Bi ₂ 0 ₃ , Bi ₂ 0 ₄ , Bi(OH) ₃ , Bi ₂ (S0 ₄ ) ₃ , Bi(N0 ₃ ) ₃ ·5H ₂ 0, (Bi0)N0 ₃ ,

(Bi0) ₂ C0 ₃ ^» 5H ₂ 0, Bi(C ₂ H ₃ 0 ₂ ) ₃ or C ₆ H ₄ C0 ₂ (Bi0)(0H). Sb ^x+ salts may include SbCI _3, Sb(N0 ₃ ) ₃ , SbCI ₅ and SbF ₃ . The elemental chalcogenide may include powders of Se, S and/or Te. Two or more chalcogenides may be used. For example, the resulting metal nanoplatelet may have the formula MS _Q SeiTei- _(Q+T), where M is a metal and Q and T range from 0 to 1. In some embodiments the nanoplatelet may have a formula including MC _n Z ₍ i- _n) , where M is a metal and y ranges from 0 to 1 . Generally, the finer the powder, the faster the Se, S and Te will be dissolved in the reaction mixture and/or be converted into the respective chalcogen anion since finer powders have a higher surface area to increase the rate of reaction and/or solubilisation. More than one chalcogen anion may react with the metal salt so that an alloyed metal nanoplatelet is formed. When more than one chalcogen anion is reacted with the metal ion, the less reactive chalcogen anion should be reacted with metal ion first otherwise. If the more reactive chalcogen anion was first reacted with the metal ion, all of the metal ions may be consumed preventing the formation of the alloy with the less reactive chalcogenide anion.

The molar ratio of metal salt to chalcogenide may be about 1 : 1 to about 1 :10. If the chalcogenide is added in excess, then the metal salt acts as the limiting reagent. This can help to ensure that all of the metal salt included in the method is converted into the metal chalcogenide nanoplatelet. In some embodiments, the molar ratio of metal salt to chalcogenide is about 1 :3. If more than one chalcogenide is used, then alloyed metal chalcogenide nanoplatelets are formed. The molar ratio of the different chalcogenides will determine the type of alloy formed. For example, for a metal chalcogenide having a formula of ZnS _x Sei- _x , X ranges from 0 and 1 . If X is 0.5, the molar ratio of S:Se is 1 :1 , so the resulting metal chalcogenide will have a formula of ZnSo 5Se ₀ 5. When more than one chalcogenide is used, the [metal]:[chalcogenide] molar ratio is determined by the total number of moles of chalcogenide, e.g. the number of moles of S and the number of moles of Se. In some embodiments, the metal chalcogenide nanoplatelet may have a formula including ZnS, ZnSe, ZnTe, ZnS _x Sei- _x , ZnSe _x Tei_ _x , CU2S, M0S2, Sb ₂ S3, B12S3 and MS _Q SeiTei-( _Q+T

Adjusting the number of moles of metal salt and chalcogenide(s) can be used to adjust the band gap of the resulting metal chalcogenide nanoplatelet. This may be important since metal chalcogenide nanoplatelets have applications in electrical applications. For example, adjusting a band gap of a material can change its electrical properties between conductors, semiconductors and non-conductors. Being able to easily adjust the band gap of the nanoplatelets means the method may be used to prepare a wide variety of nanoplatelets for a range of different applications.

An embodiment may further comprise subjecting the metal chalcogenide nanoplatelet to cation exchange. Cation exchange helps to further adjust the band gap of the nanoplatelets. Cation exchange may be performed by reacting the metal chalcogenide nanoplatelet with a second metal salt at a temperature sufficient to allow exchange of cations in the metal chalcogenide nanoplatelet. The second metal salt may include salts of Cd, Mo, W, Hg, Cu, Ag, Ni, Co, Fe, Mn, Ga, In, Sn, Ge, and Pb.

As an example, if ZnS is reacted with Cd ²⁺ in a cation exchange process, a Cd-doped ZnS nanoplatelet may be formed, or alternatively all Zn may be exchanged with Cd to form CdS nanoplatelets. The temperature sufficient to allow exchange of cations in the metal chalcogenide nanoplatelet may range from about 100 °C to about 200 °C. The specific temperature will depend on the metal chalcogenide nanoplatelet framework and the specific second metal salt.

The backbone structure of the nanoplatelet generally remains the same even after cation exchange. This means only the cation has been substituted whereas the anion framework remains unchanged. Because in some embodiments the nanoplatelets have a thickness that is substantially uniform at about 1 .40 nm, the metal chalcogenide nanoplatelet may be used as a template to synthesise further nanoplatelets having specific size properties. In this way, the metal chalcogenide nanoplatelets method may be used as a feedstock for further syntheses. This means that the disclosure also extends to a nanoplatelet formed using a method as set forth above.

An embodiment may further comprise: forming a mixture comprising a metal nanoparticle precursor and the metal chalcogenide nanoplatelets; and heating the mixture to convert the nanoparticle precursor to a nanoparticle. The formed nanoparticle may take the form of known nanoparticle shape(s). The shape(s) of the nanoparticle may include spherical shapes such as a dot, one dimensional nanorods, or two dimensional nanoplatelets. The smallest dimension of the particles may be below 60 nm. For example, the nanoparticles may have a smallest dimension greater than about 5 nm.

Forming the mixture may comprise forming a suspension of the metal nanoparticle precursor and the metal chalcogenide nanoplatelets in a solvent. The solvent may be a non-polar solvent including alkanes, alkenes, alkynes, including butane, pentane, hexane, heptane, octane and so on. Chlorinated solvents such as carbon tetrachloride, chloroform, etc. may be used. The solvent may comprise two or more different solvents, such as octadecene and hexane. The solvent may also include one or more reactants, such as reducing agents. The reactants may help to reduce the nanoparticle precursor to form the nanoparticles. For example, when the nanoparticle precursor is a metal salt, the reactants may include a reducing agent. The reducing agent may be an amine-based reducing agent. The amine-based reducing agent may be oleylamine. Prior to forming the mixture, the nanoplatelet may be isolated from the method steps mention above and purified.

The nanoparticles may be formed on a surface of the metal chalcogenide nanoplatelet. The nanoparticles may be formed on an edge surface of the

nanoplatelets. The specific location(s) where the nanoparticles are formed on the nanoplatelets may be determined by the affinity of the nanoparticles to the

nanoplatelets. For example, in some embodiments, the nanoparticles have a higher affinity to the edges of the nanoplatelet over a top or bottom surface of the

nanoplatelet. Without being bound by theory, it is thought that the top and bottom surface of the nanoplatelet have ligands (such as alkylamines used during reduction of the nanoparticle precursor to the nanoparticle) bound thereto which increases steric hindrance at the top and bottom surfaces which prevents binding of the nanoparticles to the nanoplatelets. However, if a non-amine ligand are used during nanoparticle synthesis and/or if ligands are removed from the top and bottom surfaces of the nanoplatelet prior to nanoparticle synthesis, it may be possible to promote the formation of nanoparticles on the top and bottom surfaces in addition to or in place of nanoparticle formation at edges of the nanoplatelets.

Also disclosed is a metal chalcogenide nanoplatelet, comprising a metal and one or more chalcogenides, the metal chalcogenide having a formula of MXQYTZI- _(Q+ T ₎ , wherein M is a metal, X, Y and Z are chalcogenides, and Q and T range from 0 to 1 . In some embodiments the formula may be MX _y Zi- _y , where M is a metal, X and Z are chalcogenides, and y ranges from 0 to 1 . The metal chalcogenide may be substantially free from heavy metal. The metal chalcogenide may have a thickness less than 50 nm, for example 10 nm such as less than 5 nm.

The heavy metals may be toxic heavy metals. For example, the metal chalcogenide may be free from at least one of Cd, Pb and Hg. Nanoplatelets comprising heavy metals poses a substantial threat to human health and the environment. Nanoplatelets not containing heavy metals may be of commercial significance, especially those that can be synthesised on a large scale.

Semiconductor quantum dots (containing zero-dimension spherical dots, one- dimension nanorods/nanowires, and 2D nanoplatelets) are inorganic crystals made of semiconductor materials that are small enough to exhibit quantum confinement effect. When the size of quantum dots is smaller than two times of their Bohr exciton radius (intrinsic property of semiconductors), the electron and hole are squeezed in a small space so the band gap of the semiconductor is inversely proportional to its size. 2D semiconductor nanoplatelets are a type of material that the quantum confinement effect only exhibits on the thickness direction, and they show size dependent properties only when their thickness is smaller than two times of their Bohr exciton radius. For this reason, it can be desirable to have a thickness less than two times their Bohr exciton radius. The Bohr Exciton radius depends on the composition of the nanoplatelet. For example, the Bohr exciton radius of ZnS, ZnSe and ZnTe are 2.8 nm, 4.5 nm, 5.6 nm, respectively. Therefore zinc chalcogenide nanoplatelets show size dependent properties when the thickness is smaller than 5.8 nm for ZnS, 9 nm for ZnSe, and 1 1 .2 nm for ZnTe. This may make it desirable to have the thickness of the nanoplatelet be less than 50 nm such as 10 nm, but more preferably the thickness may be less than 5 nm. In some embodiments, the thickness is less than about 3 nm or 2 nm, and in some embodiments the thickness is about 1 .40 nm, such as 1 .39 nm. In some embodiments, the thickness of the nanoplatelet is determined by the metal and chalcogenides (e.g. the atomic radii of the elements used to form the metal chalcogenide may determine the nanoplatelet thickness).

The thickness of nanoplatelet is generally an average of a size distribution from a population of nanoplatelets. Since the band gap is inversely proportional to the thickness of the nanoplatelet, having a thickness with a uniform size i.e. narrow size distribution may be advantageous. In some embodiments, the thickness of the nanoplatelet is substantially uniform at about 1 .40 nm, such as 1 .39 nm. For example, the thickness of the nanoplatelet may have a variation of about 5%.

The lateral dimensions of the nanoplatelet also have an average size calculated from a size distribution based on a population of nanoplatelets. In some embodiments the lateral dimensions of the metal chalcogenide may range from about 5 to 100 times, such as from about 15 to 40 times, the thickness of the nanoplatelet. Because the band gap of the nanoplatelet is independent of the lateral dimensions of the nanoplatelet, the specific dimensions and size distribution of the lateral dimensions may vary without affecting the band gap. However, it can be advantageous to have a population of nanoplatelets with similar lateral dimensions. The lateral dimensions of the metal chalcogenide may be less than 100 nm. For example, the lateral dimensions may range from about 30 nm to about 80 nm for a population of nanoplatelets having a thickness of about 1 .40 nm. The nanoplatelet may be made using colloidal chemical synthetic methods using a surfactant. An organic surfactant can bind tightly on the surface of

nanoparticles to prevent them from growing into big particles and forming aggregates.

If a surfactant is tightly bound to an end surface of the nanoplatelet, then growth in the lateral dimensions may be prevented since end the faces of the nanoplatelet may be prevented from reacting with additional metal salts and chalcogen anions. Conversely, if the surfactant is only loosely bound to the end surface of the nanoplatelet, then the end face of the nanoplatelet may react with metal salts and chalcogen anions in an uncontrolled manner leading to uncontrolled growth. Therefore, the type of surfactant may be selected so as to provide favourably growth kinetics i.e. provide a favourably equilibrium between the end faces of the nanoplatelet being bound to the surfactant and reacting with metal salts and chalcogen anions. If loose binding of the surfactant only occurs at an end of the nanoplatelet, then width and thickness of the nanoplatelet may remain the same as the length of the nanoplatelet increases. This type of growth mechanism may be advantageous since the length of the nanoplatelet may be determined by the reaction time.

The nanoplatelet may be substantially rectangular in shape. The term “substantially rectangular in shape” is to be interpreted broadly to mean that on 2D plane a length of one dimension is larger than the other i.e. X>Y or X<Y. For example, the nanoplatelet may have a length of about 13 nm and a width of about 4.5 nm, or a length of about 58 nm and a width of about 39 nm. The term“substantially rectangular in shape” is also to be interpreted broadly to include non-perpendicular shapes such as a rhombus, trapezoid and other shapes where one dimension is larger than another in a 2D plane, as well as morphologies including rounded and other non-linear shapes. For example, ends of a rectangular nanoplatelet may be rounded.

The nanoplatelet may be substantially square in shape. The term“substantially square in shape” is to be interpreted broadly to mean that on 2D plane a length of one dimension is the same as the other i.e. X=Y, and can include morphologies including rounded and other non-linear shapes. For example, the nanoplatelet may have a length and width ranging from about 5 nm to about 120 nm, or greater than 120 nm. The nanoplatelet may have shapes other than rectangular and square.

The metal chalcogenide may have a wurtzite crystal (i.e. hexagonal) structure. The metal chalcogenides may have a cubic structure. The metal chalcogenide may comprise two or more metals. M may be Zn, Cu, Mo, Bi, Ag, Hg and/or Sb. The two or more metals may be incorporated into the metal chalcogenide during synthesis, or they may be added into the metal chalcogenide in a post-synthesis modification step, such as cation exchange. When the metal chalcogenide has been subjected to cation exchange, M may also include Cd, Mo, Ag, W, Mn, Fe, Co, Ni, Ti, V, Zr, Hf, Nb, Ta, Cr, Ag, Hg and/or Cu.

In an embodiment, X, Y and Z may include S, Se and/or Te. As Q and T for formula MX _Q Y _T Z _I-(Q+T) , and y for formula MX _y Z _(i-y) , ranges from 0 to 1 , there may be many stoichiometric ratios of X, Y to Z. For example, the stoichiometric ratio of X to Z may be X09Z01, X085Z015, X065Z035, X04sZ052, X034Z066 and X009Z091. It should be appreciated that these ratios are exemplary only and that the stoichiometric ratio of X:Y:Z may include any integer value ranging from 0 to 1 . A formula of the metal chalcogenide may include ZnS, ZnSe, ZnTe, ZnS _x Sei- _x , ZnSe _x Tei- _x , CU ₂ S, M0S ₂ , Sb2S3 and Bi ₂ S ₃ in some embodiments. In other embodiments, the formula includes MX _Q Y _T Z _I -

The nanoplatelet may further comprise one or more types of nanoparticles bound to a surface of the nanoplatelet. For example, nanoparticles may be bound to edges of the nanoplatelet. This may form a nanoplatelet having top and bottom faces substantially free from nanoparticles. In an embodiment, the nanoparticles may include mono-metal, bi-metal, tri-metal or greater metal nanoparticles. For example, the nanoparticles may include those formed from Fe, Co, Ni, Au, Pt, Pd, Ru, Rh, Ir, CoFe, CoNi, FeNi, PtFe, PtCo, PtNi, PtAu, PtPd, PtRu, PtRh, Ptlr, PdFe, PdCo, PdNi, PdAu, PdRu, PdRh, Pdlr, AuFe, AuCo, AuNi, RuFe, RuCo, RuNi, RuAu, RuRh, Rulr, RhFe, RhCo, RhNi, RhAu, Rhlr, IrFe, IrCo, IrNi, IrAu, PtPdFe, PtPdCo, PtPdNi, Fe-ZnSe, Co- ZnSe, Ni-ZnSe, Pt-ZnSe, Pd-ZnSe, Ir-ZnSe, PtNi-ZnSe, PtPdCo-ZnSe, Co-CdS, Au- CdS, Pt-CdS, Pd-CdS, PtNi-CdS, PtPdCo-CdS, Pt-CdSe, Pd-CdSe, PtNi-CdSe and/or PtPdCo-CdSe.

Given the nanoplatelets can be semiconductors or even conductors, they may be used in optoelectronic, electrical and biomedical applications, such as light-emitting diodes, phosphors, backlight for liquid crystal displays, smart phones, and TVs, photodetectors, solar cells, catalysis, sensors and/or piezoelectrical devices.

Therefore, the disclosure also extends to an electronic device comprising the nanoplatelet as set forth above.

In one embodiment there is provided a method of forming a photodetector. The method may comprise depositing the nanoplatelets as set forth above onto an electrode. The step of depositing the nanoplatelets onto the electrode may comprise depositing a solution containing the nanoplatelets onto the electrode and subsequently evaporating the solution. The solution may be formed from a non-polar solvent including alkane, alkenes, and alkynes and/or substituted analogues thereof including halogenated alkanes. In an embodiment the solvent is chloroform, pentane, benzene, octane, decane, dimethyl ether, dichloromethane, hexane and/or toluene. The solution may be deposited onto the electrode using spin coating or other suitable surface modification processes. A concentration of the nanoplatelet solution may range from about 1 mg ml· ¹ to about 50 mg ml· ¹ . Evaporation of the solution may comprise heating the electrode and/or passing a stream of air or gas over the solution. The resulting deposited nanoplatelets may be annealed. For example, annealing may comprise heating the electrode to 300 °C in an inert atmosphere such as nitrogen. Prior to deposition of the nanoplatelets onto the electrode, the electrode may be cleaned, for example by surface treatment. Cleaning may include sonication in solvent(s) and or plasma treatment.

Brief description of figures

Embodiments will now be described by way of example only with reference to the accompanying non-limiting figures.

Figure 1 shows a one-pot ultra-large scale synthesis under ambient conditions and electron microscope images of zinc chalcogenide nanoplatelets. Figure 1 a shows a reaction setup and Figure 1 b shows a crude solution of ZnSe nanoplatelets. Figures 1 c and 1d show photographs of dried a powder sample showing a Petri dishes containing 19.9 g ZnSe nanoplatelets. A US quarter dollar coin is given in 1 c for comparison. Figure 1 e, H and 1 g shows transmission electron microscope (TEM) images of ZnS, ZnSe, and ZnTe nanoplatelets, respectively. Particles with large contract in g are tellurium metal. Insets in Figures 1 e-1 g show TEM images of nanoplatelets perpendicular aligned on the surface of amorphous carbon films. Scale bars of Insets in e-g are 5 nm. Figures 1 h, 1 i and 1j show high angle annular dark fieldscanning transmission electron microscopy (HAADF-STEM) images of ZnS, ZnSe and ZnTe, respectively. Hollow feature and patches of nanoplatelets are highlighted by dashed circles in Figure 1 i. Dots with large contrast in Figure 1j correspond to tellurium metal particles.

Figure 2 shows structural characterizations of ZnSe nanoplatelets. Figure 2a, shows a HRTEM image and Figure 2b shows a FFT images of a selected area in a marked by a red rectangle, reveal the crystallographic relation. The arrow in Figure 2a indicates the orientation of the long axis of ZnSe nanoplatelets is parallel to the crystallographic c axis of the hexagonal wurtzite structure. Figure 2c shows SAED pattern of ZnSe nanoplatelets, shows a set of distinct rings of a wurtzite hexagonal phase. Figure 2d is an FFT reconstructed image corresponding to a. Figure 2d shows AFM image (top) and the cross-section height analysis (bottom) along the long direction of an individual ZnSe nanoplatelet. Figure 2e shows XRD patterns (top) of ZnSe nanoplatelets which indicates that all diffraction peaks shift to large angle with respect to the standard XRD patterns (bottom) for wurtzite ZnSe. The standard XRD pattern for hexagonal ZnSe is presented for reference. Figure 2f is the schematic illustration of ZnSe nanoplatelets, revealing the predominant basal (1 120) facet, the side (2110) and (0002) facets.

Figure 3 shows mechanistic study of the formation of ZnSe nanoplatelets.

Figure 3a is the schematic formation illustration of ZnSe nanoplatelets, starting from ZnSe nanowires with lamellar structure. Figures 3b-e show TEM images and Figures 3f-i show HAADF-STEM images. Figures 3b and 3f show particles prepared after 2 minutes at 150 °C yielding ZnSe nanowires with lamellar structure. Figures 3c and 3g show particles prepared after 4 minutes at 150 °C yielding a mixture of fused nanowires and nanoplatelets with a small lateral size. Figures 3d and 3h show particles prepared after 30 minutes at 150 °C yielding ZnSe nanosheet with undefined shapes. Figures 3e and 3i show particles prepared after 2 h at 150 °C yielding monodisperse ZnSe nanoplatelets. All scale bar is 50 nm. Figure 3j shows thickness and length of nanoplatelets as a function of reaction time at 150 °C showing the switching of thickness of nanoplatelets from 0.9 nm to 1 .4 nm. Figure 3k show absorption spectra of ZnSe nanocrystals at different growth stages, showing the red shift of the absorption onset, in consistence with the increase of the thickness of the nanocrystals. Spectra 1 - 4 correspond to the particles prepared in Figures 3b-3e, respectively.

Figure 4 shows density function theory calculation of nanoplatelets. Figure 4a shows surface energy as a function of the chemical potential Am _å, of ZnSe slabs surface. Inset in Figure 4a shows the zoom-in of the highlighted area marked by dashed circle. Figure 4b shows a nanowire model showing the thickness of 0.99 nm. Figure 4c1 shows horizontal type oriented fusion of nanowires along [1 100] direction. Figure 4c2 shows ladder type oriented fusion of nanowires along [Ϊ 100] direction. Figure 4c3 shows oriented fusion along the [1 1 2 0] direction. Figures 4d1 and 4d2 show diffusion and nuclei construction of three additional layers along [11 2 0] and [1 Ϊ00] direction, respectively. Figure 4e shows formation energies of diffusion and nuclei construction along [11 2 0] and [1 Ϊ00] direction for three additional layers.

Figure 4f shows diffusion and nuclei construction forming (1010) and (1 Ϊ 00) terminated surfaces.

Figure 5 shows band gap tuning of nanoplatelets. TEM images of nanoplatelets a) ZnSeo ₂₅ So 75, b) ZnSeo ₂₅ So 75 (HAADF-STEM image), c), ZnSeo ₂₅ So 75. d)

ZnTeo 5Seo 5, e) CdS. F) CU ₂ S, g) M0S _2, h) Sb ₂ S3, i) CU ₂ S, and j) B1 ₂ S3. a, b, c, d, g, h, i, and j were synthesized using a wet chemical route similar to that for ZnSe

nanoplatelets e, f and g were obtained by a cation exchange reaction. Insets in a-j show the photographs of nanoplatelets dispersed in toluene solution. All the scale bars are 50 nm. k, absorption spectra of nanoplatelets. 1 -12 correspond to ZnS,

ZnSeo 25So 75, ZnSeo 25So 75, CdS, ZnSe, ZnTeo sSeo s, ZnTe, M0S2, CU2S, Sb2S3, B12S3, and Cu ₂ S, respectively, demonstrating band gap tuning has been achieved successfully.

Figure 6 shows one-pot ultra-large scale synthesis of ZnS nanoplatelets. Figure 6a shows reaction setup. Figure 6b shows crude solutions. Figures 6c and 6d show photographs of a dried sample, showing the Petri dish containing 15.9 g ZnS nanoplatelets. A US quarter dollar coin is given in Figure 6c for comparison.

Figure 7 shows thermal gravimetric analysis-differential thermal analysis (TGA- DTA) of nanoplatelets. Figure 7a shows ZnS. Figure 7b shows ZnSe. Each DTA plot contains two scans: the first scan starts from room temperature under N ₂ and finishes at 743 °C, then system cools down to a temperature below 200 °C, from which the second scans starts under air and finishes at 743 °C. Endothermic peaks are marked by dashed circles on DTA curves.

Figure 8 shows electron microscopy and sizing characterization of

nanoplatelets. Figure 8a and 8b show, respectively, TEM and HAADF-STEM images of ZnS nanoplatelets. Figures 8c and 8d show, respectively, TEM and HAADF-STEM images of ZnSe nanoplatelets. Figures 8e and 8f show, respectively, size histograms of width and length of nanoplatelets.

Figure 9 shows electron microscopy images of ZnTe nanoplatelets. Figures 9a to 9d shows, respectively, TEM, HRTEM image of a spherical tellurium particle highlighted by dashed circle in a Fast Fourier transform (FFT) of 8b, (lattice place spacings extracted from FFT match rhombohedral tellurium), and 8d HAADF-STEM image of ZnTe nanoplatelets.

Figure 10 shows HAADF-STEM image of ZnSe nanoplatelets. Hollow feature and patches of nanoplatelets are highlighted by white dashed circles, indicating nanoplatelets may form from small nanoplatelets through oriented attachment.

Figure 11 . HRTEM and FFT of selected areas of twisted ZnSe nanoplatelets. Figure 1 1 a shows HRTEM images of twisted ZnS nanoplatelets. Inserts in Figure 11 a show FFT of selected areas, as marked by rectangles, revealing the crystallographic relations. The presence of two sets of diffraction patterns are marked by arrows in top inset, revealing different orientation of the twisted lattice planes of an individual nanoplatelet. Figure 11 b shows schematic illustration of twisted nanoplatelets associated with Figure 11 a. Figure 12 shows elemental maps of zinc chalcogenide nanoplatelets. HAADF- STEM images and element mapping of a-c ZnS, d-f ZnSe, and g-i ZnTe nanoplatelets.

Figure 13 shows EDX spectrum of ZnS nanoplatelets.

Figure 14 shows EDX spectrum of ZnSe nanoplatelets.

Figure 15 shows EDX spectrum of alloyed ZnS _x Sei- _x nanoplatelets.

Figure 16 shows EDX spectrum of ZnTe nanoplatelets.

Figures 17a-17c show HRTEM and FFT of selected areas of ZnSe

nanoplatelets. Inserts show FFT of selected areas, as marked by rectangles, revealing the crystallographic relations. White dashed lines in Figure 17c are used to guide the boundary of each individual ZnSe nanoplatelet.

Figure 18 shows structural characterizations of ZnS nanoplatelets. Figure 18a shows a HRTEM image of ZnS nanoplatelets. Figure 18b shows a FFT image of a selected area in a marked by a rectangle in Figure 18a, revealing the crystallographic relation. The arrow in Figure 18a indicates the orientation of the long axis of ZnS nanoplatelets is parallel to the crystallographic c axis of the hexagonal wurtzite structure. Figure 18c shows SAED pattern of ZnS nanoplatelets. Figure 18d shows a XRD pattern (top) of ZnS nanoplatelets indicating all diffraction peaks shift to higher angles with respect to the standard XRD patterns (bottom) for wurtzite ZnS.

Figures 19a-19d show HRTEM and FFT of selected areas of ZnS nanoplatelets. Inserts in each Figure shows FFT of selected areas, as marked by rectangles, revealing the crystallographic relations.

Figure 20 shows AFM images and height analyses of individual zinc

chalcogenide nanoplatelets. Figure 20a: ZnS. Figure 20b: ZnSe.

Figure 21 a and 21 b show, respectively, TEM and HAADF-STEM images of the sample obtained after the reaction evolved for 2 min. at 150 °C. Inset in Figure 22a is HRTEM image of bundled wires showing the width of each wires is 0.99 nm.

Figure 22a and 22b show, respectively, TEM and HAADF-STEM images of the sample obtained after the reaction evolved for 4 min. at 150 °C. Figure 22a shows the product contains both fragmented belts and sheets with small lateral sizes, with the former of being the dominant phase. Figure 22b confirms the elongated belts formed from longitudinal fused wires break down into sheet.

Figures 23a-23e shows mechanistic study of the formation of ZnS nanoplatelets using TEM images of ZnS nanoparticles: Figure 23a, 10 minutes at 90 °C yielding bundled ZnS nanowires; Figure 23b, 1 minute at 110 °C yielding a mixture containing fragmented ZnS belts and sheets; Figure 23c, 10 minutes at 1 10 °C yielding ZnSe sheets with a small lateral size; Figure 23d, 30 minutes at 170 °C yielding ZnSe nanoplatelets with a large lateral size; Figure 23e, 2 hr at 170 °C yielding

monodisperse ZnS nanoplatelets. Figure 23f shows absorption spectra of ZnS nanocrystals at different growth stages, showing the redshift of the absorption onset, in consistence with the increase of the thickness of the nanocrystals. Spectra 1 -5 in Figure 23f correspond to the ZnS nanoplatelets imaged in Figures 23a)-e), respectively.

Figure 24 shows the growth direction of ZnSe nanoplatelets. Perpendicular to the surface of the paper is the [0002] direction with paralleling to the c-axis. Zinc and selenium atoms are in black and white, respectively. Figure 25 shows the slab model structures of (11 20), (1010)A/(1 100)A and (1010)B/(1 100)B, Zn-terminated (0002) and Se-terminated (0002 ) surfaces. Zinc and selenium atoms are in black and white, respectively.

Figure 26 shows the optimized structures of alkylamines ligand and ZnSe monomer adsorption on the different sites of exposed surfaces of ZnSe nanostructure. Zinc and selenium atoms are in black and white, respectively.

Figure 27 shows the upper panels show the 3D view of nanowire model, the middle panels give side views of the two equivalent structures of nanowire models, and the under panels give the optimized structures of two equivalent structures of nanowire models. Zinc and selenium atoms are in black and white, respectively.

Figure 28a and 28b shows, respectively, ZnSe monomer and octylamine binding behaviour with the (1 1 20), and (1 Ϊ00)B surface of nanowire. Zinc and selenium atoms are in black and white, respectively.

Figures 29a to 29i shows, respectively, the optimized structures referring to Figure 4c1 , Figure 4c2, Figure 4c3, additional seven layer of Figure 4d1 , additional eight layer of Figure 4d1 , additional nine layer of Figure 4d1 , additional seven layer of Figure 4d2, additional eight layer of Figure 4d2 and additional nine layer of Figure 4d2. Zinc and selenium atoms are in darker and lighter, respectively.

Figure 30 shows ZnSe nanoplatelets synthesized using different alkylamine solvents. Figure 30a shows structures of alkylamines, hexylamine (0 ₆ Hi ₅ N), octylamine (CsHigN), decylamine (C10H23N) and oleylamine (C18H37N). Figure 30b shows absorption spectra of ZnSe nanoplatelets with the absorption onset at 346 nm, suggesting uniform thickness of nanoplatelets. Spectra 1 -6 in Figure 30b correspond to the nanoplatelets in Figures 30c-30h, respectively. Figures 30c-30h show, respectively, TEM images of ZnSe nanoplatelets synthesized in octylamine, decylamine, oleylamine, hexylamine + oleylamine, octylamine + oleylamine, and decylamine + oleylamine. The volume ratio for the mixed amine solvent in Figures 30e-30g is 1 :1.

Figures 31 a-31 e show, respectively, TEM images of ZnSe nanoplatelets synthesized using Zn(N03)2-6H ₂ 0, Znl ₂ , Zn(ac) ₂ , ZnCI ₂ , Znl ₂ + ZnCI ₂ . The molar ratio of Znl ₂ + ZnCI ₂ in e is 1 :1. Figure 31 f shows absorption spectra of ZnSe nanoplatelets with the absorption onset at 346 nm, suggesting uniform thickness of nanoplatelets. Spectra 1-5 in Figure 31 f correspond to the nanoplatelet synthesis conditions for TEM images a)-e), respectively.

Figure 32a shows absorption spectra of ZnSe nanoparticles obtained at different temperatures, showing the conversion the formation of ZnSe nanowires from the preceding ZnSe nanoplatelets. Figure 32b shows a TEM image of ZnSe nanowires obtained at 170 °C after 60 min. Figure 32c shows a TEM image of ZnSe nanowires obtained at 220 °C after 30 min.

Figure 33 shows growth of ZnTe nanoplatelets. Figure 33a shows UV-Vis absorption spectrum (solid line) and photoluminescent spectrum (dashed line) of ZnTe magic sized nanoclusters (MSNCs) and ZnTe nanoplatelets. Figure 33b shows a schematic illustration of ZnTe nanoplatelets formed in soft colloidal templates from (ZnTe)i3 and (ZnTe)34 MSNCs. Figures 33c-33e show, respectively, TEM image of (ZnTe)i3 (120 °C, 2 hours), (ZnTe) ₃ 4 MSNCs (200 °C, 2 minutes), and ZnTe nanoplatelets (200 °C, 30 minutes). Figure 34 shows structural characterizations of ZnTe nanoplatelets. Figure 34a shows a TEM image of single-layered ZnTe nanoplatelets. Figure 34b shows a

HRTEM image of an individual ZnTe nanoplatelet in Figure 34a. Figure 34c shows a FFT of the HRTEM image of Figure 34b. Figure 34d shows a SAED pattern of ZnTe NPLs. Figure 34e shows a TEM images of ZnTe nanoplatelets standing on their edges; inset shows a zoom-in of a single nanoplatelet. Figure 34f shows a AADFSTEM image and STEM-EDX elemental maps of ZnTe nanoplatelets.

Figure 35 shows a ZnSe nanoplatelet-based photodetector. Figure 35a shows a microscopic image showing the active area of one photodetector. Figure 35b shows Current- Voltage characteristics of one photodetector measured at dark (squares), with 365 nm (500 pW cm ^-2 ; circles) and 254 nm (650 pW cm ^-2 ; triangles) UV light irradiation, respectively. Figures 35c and 35d show, respectively, transient responses of the photodetector with light ON and OFF and 100 V biasing for 365 nm UV light and 254 nm UV light.

Figures 36a-36b show transient responses of the ZnSe photodetector at 365nm, and Figures 36c-36d show transient responses of the ZnSe photodetector at 254nm.

Figure 37 shows optical spectroscopy of zinc chalcogenide nanoplatelets.

Figure 37a shows TEM and HAADF-STEM (inset) images of ZnSo 75Seo 25

nanoplatelets. Figure 37b shows TEM and HAADFSTEM (inset) images of

ZnS ₀ 5oSeo 5o nanoplatelets. All scale bars are 50 nm. Figure 37c shows a comparison of absorption (solid lines) and fluorescence emission (dash-dotted lines) spectra of zinc chalcogenide nanoplatelets. The light hole-electron and heavy hole-electron transitions are marked as Ih-e and hh-e, respectively. Figure 37d shows normalized fluorescence emission spectra of zinc chalcogenide nanoplatelets showing the tunability of the emission wavelengths from 296 nm to 349 nm. Figure 37e shows normalized fluorescence emission spectra of ZnSe nanoplatelets for different temperatures. Figure 37f shows fluorescence lifetimes of zinc chalcogenide nanoplatelets at room

temperature.

Figure 38 shows fluorescence emission spectra of nanoplatelets at room temperature (RT) and 77 K.

Figure 39 shows fluorescence lifetime of zinc chalcogenide nanoplatelets at 77 K. The fluorescent emission lifetime curve for ZnSo 75Seo 25 (toluene + TBP) at 77 K show a bi-exponential decay, with t1 = 1 .76 ns and t2= 3.92 ns. The curves indicated by the arrow are in order: ZnS, ZnSe, ZnSo 50Seo 50, and ZnSo 75Seo 25.

Figure 40 shows a comparison of XRD patterns of nanoplatelets before and after ligand exchange.

Figure 41 shows a comparison of absorption spectra of nanoplatelets before and after ligand exchange.

Figure 42 shows an XRD pattern of bimetal Ptlr-ZnS nanoplatelets.

Figure 43 shows an XPS spectrum of bimetal Ptlr-ZnS Nanoplatelets.

Examples

Embodiments will now be described with reference to the following non-limiting examples.

Materials Zinc nitrate hexahydrate (Zh(N0 ₃ ) ₂ ·6H ₂ 0, 98%), zinc chloride (ZnCI ₂ , 99.999%), zinc acetate dihydrate [Zn(CH ₃ C00) ₂ -2H ₂ 0, 99.999%], zinc iodide (Znl ₂ , >98%), cadmium(ll) oxide (CdO, >99.99%), lead acetate trihydrate [Pb(CH ₃ C00) ₂ -3H ₂ 0, >99.999%], molybdenum(V) chloride (MoCI ₅ , 95%), copper(l) chloride (CuCI,

>99.995%), BiCIs (99.999%), SbCI ₃ (99,95%), selenium powder (>99.5%), sulfur powder (99.998%), tellurium powder (99.8%), hexylamine (99%), octylamine (>98%), decylamine (95%), dodecylamine (98%), oleylamine (97%), oleic acid (tech., 90%), 1 - octadecene (ODE, tech., 90%), trioctylphosphine (TOP, 97%), superhydride

(LiBH(CH ₂ CH ₃ )3) solution in THF (1 M), chloroform (99% anhydrous) and methanol (99.8% anhydrous) were purchased from Sigma Aldrich. All chemicals were used without further purification.

Synthesis of nanoplatelets

Firstly, all the syntheses were carried out under inert atmosphere using standard Schlenk techniques. Then the same syntheses were also conducted under ambient conditions in order to elucidate if the syntheses of nanoplatelets are sensitive to oxygen and moisture. The quality of products obtained under ambient conditions is the same as that obtained using inert atmosphere using standard Schlenk techniques. Therefore, in the following experiments, all the syntheses were carried out under ambient conditions unless otherwise stated.

ZnSe nanoplatelet synthesis

Monodisperse ZnSe nanoplatelets were synthesized using a one-pot synthetic method. In a typical synthesis, 0.6 mmol ZnCI ₂ and 1 .8 mmol selenium powder were mixed with 10 ml_ decylamine and 20 mL oleylamine in a 50 mL three-neck flask. The mixture was heated to 110 °C and kept at this temperature for 30 min. Then, the temperature was held at 150-170 °C for a desired reaction time (refer Table 1 for details). The reaction was quenched by removal of the heating mantle. After cooling to room temperature, the crude reaction mixture was diluted with chloroform. 2 mL trioctylphosphine was added in order to extract the unreacted selenium. Methanol was added to precipitate the nanoparticles with the aid of centrifugation.

Table 1. Summary of the syntheses of ZnSe nanoplatelets

Zinc Source Primary Amines Reaction Reaction First Thickness

Temperature Time exciton

peak

Hexylamine 120 °C 6 h 346 nm 1 .4 nm

Octylamine 170 °C 2 min-6 h 346 nm 1 .4 nm

Decylamine 170 °C 2 h-6 h 346 nm 1 .4 nm

Zn(N0 ₃ ) ₂ -6H ₂ 0 Oleylamine 170 °C 2 h-10 h 346 nm 1 .4 nm

Hexylamine+Oleylamine 120 °C 2 h-12 h 346 nm 1 .4 nm Octylamine+Oleylamine 170 °C 2 min-6 h 346 nm 1 .4 nm Decylamine+Oleylamine 170 °C 10 min-1 .5h 346 nm 1 .4 nm ZnCI ₂ Octylamine+Oleylamine 170 °C 2 h- 6 h 346 nm 1 .4 nm

ZnAc ₂ -2H ₂ 0 Octylamine+Oleylamine 170 °C 10 min-10 h 346 nm 1 .4 nm

Zn l ₂ Octylamine+Oleylamine 170 °C 10 min-6 h 346 nm 1 .4 nm

ZnCI ₂ + Znl ₂ Decylamine+Oleylamine 170 °C 10 min-6 h 346 nm 1 .4 nm

Up-scale synthesis of ZnSe nanoplatelets

The synthetic procedure for up-scale synthesis of ZnSe nanoplatelets was the same as that for the typical synthesis of ZnSe nanoplatelets described above. In this synthesis, Zh(N0 ₃ ) ₂ ·6H ₂ 0 (100 mmol, 29.7 g) and selenium powder (300 mmol, 23.7 g) were mixed with 120 ml_ octylamine and 240 ml_ oleylamine in a 500 ml_ flask using an oil bath as the heating source (Figure 1 a).

ZnS nanoplatelet synthesis

Monodisperse ZnS nanoplatelets were synthesized using a one-pot synthetic method. Typically, Zn(N0 _3j2 -6H _2Q (0.3 mmol, 89 mg) and sulfur (0.9 mmol, 28 9 mg) were mixed with 5 ml decylamine and 10 mL oleylamine in a 50 mL three-neck flask. The mixture was heated to 110 °C and kept at this temperature for 30 min. The temperature was then held at 170 °C for a desired reaction time, before removal of the heating mantel and cooled down to room temperature. The product solution was diluted in chloroform, and the nanoplatelets were precipitated by adding methanol with the aid of centrifugation.

Up-scale synthesis of ZnS nanoplatelets

The synthetic procedure for up-scale synthesis of ZnS nanoplatelets was the same as that for the typical synthesis of ZnS nanoplatelets described above. In this synthesis, Zh(N0 ₃ ) ₂ ·6H ₂ 0 (100 mmol, 29.7 g) and sulphur powder (300 mmol, 9.6 g) were mixed with 120 mL decylamine and 240 mL oleylamine in a 500 mL flask using an oil bath as the heating source (Figure 5a). Alloyed ZnS _x Sei- _x nanoplatelet synthesis

The synthetic procedure for alloyed ZnS _x Sei- _x nanoplatelets was the same as that for ZnS nanoplatelets, except a mixture of chalcogens containing selenium (0.45 mmol, 35.5 mg) and sulfur (0.45 mmol, 14.5 mg) were used.

ZnTe nanoplatelet synthesis

The synthesis of ZnTe nanoplatelets was carried out under an inert atmosphere using standard Schlenk techniques. In a typical synthesis, 0.5 M tellurium stock solution was prepared by dissolving 2 mmol tellurium powder in 4 ml_ trioctylphosphine at 200 ^° C. Zh(N03)2 6H ₂ 0 (0.3 mmol, 89 g) were mixed with 5 ml decylamine and 10 ml oleylamine in a 50 ml_ three-neck flask. The mixture was degassed and refilled with N ₂ for three times at room temperature and then heated to 1 10 °C and kept at this temperature for 30 min. The tellurium stock solution (0.5 M, 0.6 mL) was mixed with the superhydride solution (1 M, 0.42 mL) in the glove box before injection of tellurium precursor into the zinc solution at 160 °C. After the injection, the temperature was increased to 170 ^° C and held this temperature for 6 h, before removal of the heating mantel and cooled down to room temperature. The product solution was diluted in chloroform, and the nanoplatelets were precipitated by adding methanol with the aid of centrifugation.

Alloyed ZnSe _x Tei- _x nanoplatelet synthesis

The synthesis of alloyed ZnSe _x Tei_ _x nanoplatelets was carried out under an inert atmosphere using standard Schlenk techniques. In a typical synthesis, 0.2 M selenium stock solution was prepared by dissolving 2 mmol (159 mg) selenium powder in 10 mL oleylamine at 220 °C. Zn(NG ₃ ) ₂ -6H _2Q (0.3 mmol, 89 mg) were mixed with 5 mL decylamine and 10 mL oleylamine in a 50 mL three-neck flask. The mixture was degassed and refilled with N ₂ for three times at room temperature and then heated to 110 °C and kept at this temperature for 30 min. At 60 °C, the Se-oleylamine solution (0.2 M, 1.5 mL) mixed with superhydride solution (1 M, 0.42 mL) was injected into the flask. Then the solution was heated up to 170 °C. The tellurium stock solution (0.5 M, 0.6 mL) was mixed with the superhydride solution (1 M, 0.42 mL) in the glove box before injection of tellurium precursor into zinc solution at 170 °C. As soon as the temperature of the reaction solution reached 170 °C, 0.6 mL tellurium precursor (0.5 M) was swiftly injected into the flask and the reaction solution was held at 170 °C for 2.5 h, before removal of the heating mantel and cooled down to room temperature. The product solution was diluted in chloroform, and the nanoplatelets were precipitated by adding methanol with the aid of centrifugation.

Synthesis of other 2D nanoparticles

Preparation of sulphur oleylamine stock solution: 0.1 M sulphur stock solution was prepared by dissolving 3 mmol (0.096 g) sulphur powders into 30 mL oleylamine at 180 °C.

Sb ₂ S ₃ : In a typical synthesis, 0.1 mmol (22.8 mg) SbCI ₃ was mixed with 5 mL oleylamine in a three-neck flask. Then the mixture was heated up to 1 10 °C and kept at this temperature for 0.5 hr. At 160 °C, 2.25 mL 0.1 M sulphur stock solution was swiftly injected into the flask. Then the mixture was heated up to 230 °C and kept at this temperature for 20 min, before removal of the heating mantel and cooled down to room temperature.

Cu _? S: In a typical synthesis, 0.2 mmol (19.8 mg) CuCI was mixed with 10 mL oleylamine in a three-neck flask. Then the mixture was heated up to 1 10 °C and kept at this temperature for 0.5 hr. At 160 °C, 3 mL 0.1 M sulphur stock solution was swiftly injected into the flask. Then the mixture was heated up to 200 °C and kept at this temperature for 5-20 min, before removal of the heating mantel and cooled down to room temperature.

B1 ₂ S ₃ : In a typical synthesis, 0.2 mmol (63 mg) BiCI ₃ was mixed with 10 mL oleylamine in a three-neck flask. Then the mixture was heated up to 1 10 °C and kept at this temperature for 0.5 hr. At 160 °C, 4.5 mL 0.1 M sulphur stock solution was swiftly injected into the flask. Then the mixture was kept at this temperature for 30 min to 2 hr, before removal of the heating mantel and cooled down to room temperature.

Cation exchange reaction

From ZnS to MoS ₂

0.2 M MoCis-oley!amine stock solution was prepared by dissolving MoCi ₅ (0.68 g, 2.5 mmol) in oleylamine (12.5 mL) at 200 °C. Purified ZnS nanoplatelets (containing 3.3 x 10 ^~5 mol Zn ^2* ) was dissolved in anhydrous chloroform and then the solution was injected into 5 mL oleylamine in a 25 mL flask sealed with a septum. The mixture was degassed at 1 10 °C for 30 min, and then the mixture was heated to 150 °C. When the temperature was stabilized, 0.5 mL MoGb-oleylamine stock solution stock solution (0.2 M) was injected into the flask while stirring. The colourless solution changed to dark brown at about 150 °C in 10 min. The solution was cooled to room temperature.

From ZnS to CdS

0.1 M cadmium oleate stock solution was prepared by dissolving CdO (0.32 g, 2.5 mmol) in a mixture of oleic acid (6.94 mL) and ociadecene (18 mL) at 220 °C. Cleaned ZnS nanoplatelets (containing 3.3 ^c 1 G ⁵ mol Zn ² *J were dissolved in anhydrous chloroform and then the solution was injected into 5 mL oleylamine solution in a three-neck flask. The mixture was degassed at 1 10 °C for 30 min, and then it was heated to 170 °C. When the temperature was stabilized, 1 mL cadmium oleate stock solution (0.1 M) was swiftly injected into the flask while stirring. The solution in the flask gradually changed from colourless to bright yellow and finally to yellow. After 30 min at this temperature the heating mantle was removed and the mixture was cooled to room temperature.

From ZnS to Cu _? S

0 1 M CuCI-o!ey!amine stock solution was prepared by dissolving CuCI (0 50 g, 0.5 mmol) in oleylamine (5 mL) at 110 °C. Purified ZnS nanoplatelets (containing 3.3 ^c 1 Q ^~5 mol Zn ²⁺ ) was dissolved In anhydrous chloroform and then the solution was injected into 5 mL oleylamine in a 25 mL flask sealed with a septum. The mixture was degassed at 1 10 °C for 30 min, and then it was cooled to room temperature. When the temperature was stabilized, 1 rnL CuCI~o!ey!amine stock solution stock solution (0 1 M) was injected into the flask while stirring. The colourless solution changed to dark brown immediately at room temperature.

Post-functionalisation of nanopiateiets with nanoparticles

All the syntheses were performed using standard Schlenk line techniques.

Synthesis of Metal-ZnS Nanoplatelets: M = (Fe, Co, Ni, Au, Pt, Pd, Ru, Rh, Ir)

In a typical procedure, a mixture of required amount of M(acac) ₂ , 120 mI_ of oleylamine, 8 ml_ of ODE and 1.2 mg of ZnS nanoplatelet powders in hexane (so that the molar precursor ratio Zn:M = 1 :5) were degassed at 50 °C for 30 min. Under N ₂ flow, the temperature was raised to 130 °C, and the reaction was continued for 2.5 h. The reaction was quenched by removal of the heating mantle. After cooling to room temperature, a 15 mL mixture of methanol/chloroform = 1 :1 was added. The whole product was centrifuged at 6000 rpm for 10 min, the supernatant was discarded and the precipitate of the nanoparticle-functionalised nanoplatelet was dispersed in 3 mL hexane. For the growth of Fe, Co Ni and Ir, the temperature was set to 160 °C but all other reaction conditions were the same as above.

Bimetal Mi-M _? -ZnS Nanoplatelets

In a typical procedure, a mixture of required amount of Mi(acac) ₂ , M ₂ (acac) ₂ ,120 pL of oleylamine, 8 mL of ODE and 1.2 mg of ZnS nanoplatelet powders in hexane (so that the molar precursor ratio Zn:Mi:M ₂ = 1 :5:5) were degassed at 50 °C for 30 min. Under N ₂ flow, the temperature was raised to 130 °C, and the reaction was continued for 2.5 h. The reaction was quenched by removal of the heating mantle. After cooling to room temperature, a 15 mL mixture of methanol/chloroform = 1 :1 was added. The whole product was centrifuged at 6000 rpm for 10 min, the supernatant was discarded and the precipitate of the nanoparticle-functionalised nanoplatelet was dispersed in 3 mL hexane.

Trimetal Mi-M ₂ -M ₃ -ZnS Nanoplatelets

In a typical procedure, a mixture of required amount of Mi(acac) ₂ , M ₂ (acac) ₂ ,

M ₃ (acac) ₂ ,120 pL of oleylamine, 8 mL of ODE and 1 .2 mg of ZnS nanoplatelet powders in hexane (so that the molar precursor ratio Zn:Mi:M ₂ :M ₃ = 1 :5:5:5) were degassed at 50 °C for 30 min. Under N ₂ flow, the temperature was raised to 130 °C, and the reaction was continued for 2.5 h. The reaction was quenched by removal of the heating mantle. After cooling to room temperature, a 15 mL mixture of methanol/chloroform = 1 :1 was added. The whole product was centrifuged at 6000 rpm for 10 min, the supernatant was discarded and the precipitate of the nanoparticle-funciionalised nanoplatelet was dispersed in 3 mL hexane.

Synthesis of Metal-ZnSe Nanoplatelets

The reaction procedure was the same as above for metal-ZnS nanoplatelets, except that 1 .8 mg of ZnSe nanoplatelet powders in hexane were used.

Synthesis of Metal-CdS Nanoplatelets The reaction procedure was the same as above for metal-ZnS nanoplatelets, except that 1 .8 mg of CdS nanoplatelet powders in hexane were used.

Synthesis of Metal-CdSe Nanoplatelets

The reaction procedure was the same as above for metal-ZnS nanoplatelets, except that 2.36 mg of CdSe nanoplatelet powders in hexane were used.

Bimetal Ptlr-ZnS Nanoplatelets

In a typical procedure, a mixture of required amount of Pt(acac) ₂ , lr(acac) ₃ ,120 pL of oleylamine, 8 ml_ of ODE 1 .2 mg of ZnS nanoplatelet powders in hexane (so that the molar precursor ratio Zn:Mi:M2 = 1 :5:5) were degassed at 50 °C for 30 min. Under N ₂ flow, the temperature was raised to 130 °C, and the reaction was continued for 2.5 h. The reaction was quenched by removal of the heating mantle. After cooling to room temperature, a 15 ml. mixture of methanol/chloroform = 1 :1 was added. The whole product was centrifuged at 6000 rpm for 10 min, the supernatant was discarded and the precipitate of the nanoparticle-functionalised nanoplatelet was dispersed in 3 mL hexane. An XRD pattern of bimetal Ptlr-ZnS nanoparticle-functionalised nanoplatelets is shown in Figure 42. The individual components of the Ptlr-ZnS nanoparticle- functionalised nanoplatelet are indicated in Figure 42; unlabeled peaks are attributed to ZnS. An XPS spectrum of bimetal Ptlr-ZnS nanoparticle-functionalised nanoplatelets is shown in Figure 43.

Alloy PtNi-CdSe Nanoplaielets

In a typical procedure, a mixture of required amount of Pt(acac) ₂ , Ni(acac) ₂ ,12Q pL of oleylamine, 8 mL of ODE 2.4 mg of CdSe nanoplatelet powders in hexane (so that the molar precursor ratio Zn:M1 :M2 = 1 :5:5) were degassed at 50 °C for 30 min. Under N ₂ flow, the temperature was raised to 130 °C, and the reaction was continued for 2.5 h. The reaction was quenched by removal of the heating mantle. After cooling to room temperature, a 15 L mixture of methanol/chloroform = 1 :1 was added. The whole product was centrifuged at 6000 rpm for 10 min, the supernatant was discarded and the precipitate of the nanoparticle-functionalised nanoplatelet was dispersed in 3 mL hexane.

Instrument Specifications & Sample Preparation

UV-Visible-Near IR absorption spectroscopy

UV-vis absorption spectroscopy was performed on a Perkin Elmer Lambda 35 UV/VIS Spectrometer using quartz cuvettes.

Powder X-ray Diffraction (XRD) patterns

Powder X-ray diffraction (XRD) patterns were obtained using Cu Ka (l=1 .5406 A) photons from an X'per PRO (PANalytical) X-ray diffractometer operated at 40 kV and 40 mA. The crude solutions of nanoplatelet samples were diluted in chloroform and precipitated by methanol in the aid of centrifugation. After three times of wash, the samples were dried in a glove box overnight before XRD analysis. In the case of ZnSe nanoplatelets, a small amount of trioctylphosphine was added together with methanol in order to remove the unreacted selenium powder. Samples were deposited as a thin layer on a low-background scattering silicon substrate.

Electron Microscopy Measurement

Transmission electron microscopy (TEM) was performed using a JEOL 2100 transmission electron microscope. High-resolution TEM (HRTEM), high angle annular dark field scanning transmission electron microscopy (HAADF-STEM) and STEM- Energy-dispersive X-ray spectroscopy (EDX) were performed on an FEI Titan G2 80- 200 high-resolution transmission electron microscope. Samples for electron microscopy measurement were prepared by placing a drop of nanoparticle solution in chloroform on top of a copper grid coated with an amorphous carbon film.

Thermal gravimetric analysis-differential thermal analysis

The analysis was performed on a TA Instruments SDT Q600 simultaneous DTA-TGA. Samples were prepared similar to those for XRD. Approximately 10 mg of sample was weighed into a 1 10 pL platinum crucible with a matched empty crucible as a reference. The sample was heated from ambient to 750°C at 10°C per minute in a nitrogen atmosphere flowing at 100 ml per minute, cooled to 200°C, then heated to 750°C at 10°C per minute in air flowing at 100 ml per minute. The temperature scale of the instrument was calibrated using the melting points of 99.999% indium

(156.5985°C), 99.99+% tin (231 .93°C), 99.99+% zinc (419.53°C), 99.99% silver (961 .78°C), and 99.999% gold (1064.18°C). The balance was calibrated over the temperature range used with standards provided by the instrument manufacturer. The heat flow between the pans was calibrated using a sapphire disk provided by the manufacturer. The cell constant was fine-tuned using the heat of fusion of zinc (1 13

J/g)·

Discussion

Synthesis

In comparison to the widely used standard Schlenk techniques for the preparation of colloidal nanoparticles that require moisture- and oxygen-free environment, the syntheses of ZnS and ZnSe nanoplatelets were conducted under ambient conditions using easily accessible experimental apparatus containing a hot plate, an oil bath and a reacting flask (Figure 1 a and Figure 6). For typical syntheses, zinc chalcogenide nanoplatelets were synthesized via a one-pot synthetic method using zinc salts and elemental chalcogens as the precursors, and a co-solvent containing short carbon chain amine (C ₆ -Ci2)/oleylamine mixture as both solvents and surfactants. For up-scaled syntheses, 15.9 g ZnS (Figs. 1 c, 1 d) and 19.9 g ZnSe (Figure 6c and 6d) dried powder samples were obtained after the purification of ~360 mL crude reaction solutions (Figure 1 b and Figure 6b). Thermal gravimetric analysis- differential thermal analysis (TGA-DTA) (Figure 7) on the purified powder samples reveal organic ligands contribute approximately 49 wt% of the total collected mass. This value is much higher than the ligand loss (~20 wt%) in the case of spherical nanoparticles because the flat, extended facets of 2D nanoplatelets can support more ligand interaction through better surface contact than curved facets in spherical particles. It is noteworthy that ZnS nanoparticles are more stable in air with no evident oxidization below 560 °C compared with ZnSe ones (Figure 7). Based on the TGA data, we obtained volumetric production capacities of 22.2 g/L for ZnS and 28.5 g/L for ZnSe nanoplatelets.

Analogous to Figure 1 a-1 d in the main article, Figure 6a shows the reaction setup for one-pot ultra-large scale synthesis of ZnS nanoplatelets. 15.9 g ZnS nanoplatelets (Figire 6c) were obtained after the cleaning of the reaction crude solution (360 mL) (Figure 6b).

For ZnTe nanoplatelets, the use of a reducing agent is required to form the telluride required for synthesis. A step-wise transition of the absorption spectra is observed (Figure 33a), indicating that the ZnTe nanoplatelets are evolved from

(ZnTe)i3 and (ZnTe)34 magic sized nanoclusters (MSNCs). The absorption spectra of aliquots taken at various reaction stages clearly show the sharp absorption doublets, which represents two electronic transitions with the lowest energies. As the reaction proceeded for 2 hours at 120 °C, the first aliquot was taken and its absorption spectrum (Figure 33a) presents a closely spaced doublet-like feature with two peaks at 297 and 322nm, which are typically characteristics of (ZnTe) _i3 MSNCs. At 200 °C, as the reaction proceeded for 2 minutes, the second aliquots was taken and its absorption spectrum (Figure 33a) exhibits a similar doublet-like feature but with two peaks at longer wavelengths of 362 and 396nm, matching the unique characteristics of (ZnTe) MSNCs. The absorption doublet peaks at 297 and 322nm of (ZnTe) _i3 MSNCs are assigned to the transitions from 1 S3/2(h) to 1 S(e) and from 1 S3/2(I) to 1 S(e) respectively, whereas those absorption peaks at 362 and 396nm of (ZnTe) ₃ 4 MSNCs correspond to the transitions from 2S3/2(h) to 1 S(e) and from 2S3/2(I) to 1 S(e), being analogous to CdSe MSNCs. As the reaction proceeded for 30 minutes at 200 °C, ZnTe nanoplatelets were obtained and the exciton peaks in the absorption spectrum of ZnTe nanoplatelets (Figure 33a) have the same wavelengths as those of (ZnTe) MSNCs, but clearly they are much sharper, suggesting excitons are one-dimensionally well- confined in the thickness direction of ZnTe NPLs. Both (ZnTe) _i3 and (ZnTe) ₃ 4 MSNCs do not have detectable photoluminescence. However, the room-temperature PL spectrum of the ZnTe nanoplatelets exhibits a sharp emission band with a full width at half maximum of 10 nm at 400nm (78 meV) under 365 nm excitation, thus confirming the atomically uniform flatness of the ZnTe NPLs (Figure 33a). Figure 33b presents the schematic illustration of ZnTe nanoplatelets formed stepwise in soft colloidal templates from (ZnTe)i3 and (ZnTe) ₃ 4 MSNCs. The TEM image of (ZnTe) _i3 MSNCs (Figure 33c) shows that the MSNCs with a lateral size of ~4nm are assembled in coin-like templates and stacked with each other, which is apparently different from that reported of (ZnTe) 13 MSNCs with a triangle plate shape. The thickness of self-assembled (ZnTe)i3 MSNCs is ~1.1 nm, as obtained from the densely stacked (ZnTe) _i3 MSNCs standing on their edges (Figure 33c). As the reaction proceeded for 2 minutes at 200 °C, irregular rectangles with lateral dimensions of 15-20 nm are formed by the self- assembly of (ZnTe) ₃ 4 MSNCs (Figure 33d). The absorption exciton peak at a lower energy has a profound red-shift from 322 nm to 396 nm, which corresponds to the increase of the thickness of self-assembled MSNCs from 1.2 nm to 1.5 nm, as discussed later. After the reaction evolved for 30 minutes at 200 °C, well-defined rectangle-shaped ZnTe nanoplatelets with lateral dimensions of ~15 nm x 60 nm are obtained (Figure 33e). Figure 34a shows the TEM image of single-layered ZnTe nanoplatelets. The high-resolution TEM (HRTEM) image (Figure 34b) and selected-area electron diffraction (SAED) pattern (Figure 34d) of ZnTe nanoplatelets reveal that the ZnTe nanoplatelets are single-crystalline with a hexagonal wurtzite structure. The fast Fourier transform (FFT) of the HRTEM in Figure 34b shows the crystal planes of the wurtzite ZnTe nanoplatelets, as labelled in Figure 34c. Lattice spacings (Figure 34b) extracted from FFT (Figure 34c) are 0.355nm (002) and 0.374nm (100), which are typical of wurtzite ZnTe, confirming the length direction of nanoplatelets corresponds to the c-axis of wurtzite structure. The SAED pattern in Figure 34d is well-indexed to the Miller indices of wurtzite ZnTe. Figure 34e shows that each ZnTe nanoplatelet (Figure 34e) stands on its edge, from which the thickness of a single NPL is estimated to be ~1 5nm. The upper-right inset of Figure 34e shows the good crystallinity and the uniform thickness of the ZnTe nanoplatelets. Figure 34f presents the high-angle annular dark-field (HAADF) image and STEM-EDX elemental maps of ZnTe nanoplatelets, confirming the Zn and Se elements are evenly distributed in the nanoplatelets.

Thermal property analysis of nanoplatelets

Figure 7a and 7b show TGA-DTA curves of ZnS and ZnSe nanoplatelets, respectively. In the first scan conducted in N ₂ from room temperature to 743 °C, TGA curves present a dominant weight loss of the samples (49.7% for ZnS and 48.4% for ZnSe) in the temperature range of 150-460 °C accompanied by small endothermic peaks, as marked by red circles on DTA curves. The weight loss and endothermic feature correspond to the evaporation of octylamine (boiling point 179.4 °C) and oleylamine (boiling point 350 °C) ligands. Thermal property measurements conducted in air show a second weight loss (8.8% for ZnS and 20.6% for ZnSe) in TGA curves in correlation with sharp exothermic peaks (635 °C for ZnS and 465 °C for ZnSe) in the DTA curves. Close inspections on the second weight loss in conjugation with the exothermic feature suggest that zinc chalcogenide nanoplatelets convert into ZnO due to the oxidation in air at elevated temperatures.

2ZnS + 30 ₂ 2ZnO + 2SO ₂ † (Equation 1)

Theoretical weight loss for ZnS based on Equation 1 :

I 4.1

The values of measured weight loss for ZnS and ZnSe conducted in air are 8.8% and 20.6%, being in consistent with those of the theoretical weight loss 8.3% and 22.5%, respectively. This consistency confirms that the presence of the exothermic peaks in the DTA curves corresponds to the decomposition of zinc chalcogenide nanoparticles due to oxidization. Compared with ZnSe nanoplatelets, ZnS

nanoplatelets are more stable in air, with no evident oxidization below 560 °C.

Estimated reaction yield and volumetric production capacity based on the TGA- DTA data

Theoretical yield =

Volumetric production capacity of ZnS = 15.9 *(1 -49.7%)/0.360 = 22.2 g/L.

Volumetric production capacity of ZnSe = 19.9 *(1 -48.4%)/0.360 = 28.5 g/L.

TEM and HAADF-STEM characterizations of nanoplatelets

The TEM and HAADF-STEM images of large populations of zinc chalcogenide nanoplatelets are used to illustrate the particles shown in Figures 1 e-j, are

representative of the entire samples. The sizing histograms (Figure 8e and 8f) indicate that both ZnS and ZnSe nanoplatelets have narrow size distributions.

In addition to Figure 1 i, HAADF-STEM image in Figure 10 shows hollow and patchy features, suggesting ZnSe nanoplatelets may form via oriented attachment.

TEM images of the reaction products show nearly monodisperse nanoplatelets with lateral dimensions ranging from 30 nm to 80 nm (Figures 1 e-g, 8 and 9). TEM measurements on nanoplatelets standing on their edges reveal the uniform thickness of 1 .39 nm (Insets in Figures 1 e-g). The contrast of ZnS and ZnSe nanoplatelets in HAADF-STEM images is very homogenous throughout the entire surface of nanoplatelets and they form a quite uniform flat rectangular shape (Figures 1 h-j, 8b and 8d). A close inspection of the HAADF-STEM image of ZnSe sample (Fig. 1 i,

Figure 10) reveals evident hollow feature and patches on some individual

nanoplatelets, indicating nanoplatelets may form by oriented attachment. Twisting of ZnSe nanoplatelets along the long lateral direction were also observed (see Figure 1 1), which may be attributed to the strain imposed by the ligands on the surface. The ZnTe samples are less stable in air and underwent partially oxidization after being separated from the crude reaction solution, producing tellurium metal dots decorated

nanoplatelets (Figure 1 g and 8). STEM-energy dispersive x-ray (EDX) element maps (Figure 12) of zinc chalcogenide nanoplatelets show that both Zn and chalcogen are evenly distributed throughout the 2-D structures with a roughly stoichiometric ratio of 1 :1 (Figures 13-16, Table 2).

Table 2. EDX spectroscopy analyses of zinc chalcogenide nanoplatelets.

Sample Zn at% S at% Se at% Te at% ZnS 47.3 52.7

ZnSe 45.5 54.5

ZnS _x Sei- _x 58.7 28.6 12.7

ZnTe 44.7 55.3

High resolution TEM (HRTEM) measurements (Figures 2 and 17-19) shows the nanoplatelets are crystalline. Lattice plane spacings extracted from fast Fourier transform analysis of selected areas are 0.335 nm (2Ϊ Ϊ0) and 0.318 nm (0002) for ZnSe (Figure 2a) and 0.322 (2Ϊ Ϊ0) and 0.304 nm (0002) for ZnS nanoplatelets (Figure 18), typical of the wurtzite hexagonal structure. Selected area electron diffraction (SEAD) patterns of nanoplatelets in Figure 2c and Figure 18c show a set of well- defined rings indexed to hexagonal ZnSe and ZnS, respectively, being in consistence with their respective HRTEM results. Atomic force microscope images of individual ZnS (Figures 2e and 20a) and ZnSe (Figure 20b) nanoplatelets reveal the uniform height of ~1 .39 nm, being consistent with the TEM results on nanoplatelets standing on their edges (Insets of Figures 1 e and 1 f). The X-ray diffraction patterns of ZnS (Figure 18d) and ZnSe (Figure 2f) match hexagonal wurtzite structure but clearly show a slightly shift to higher angles in comparison with each individual standard XRD pattern. This indicates that the lattices of ZnS and ZnSe nanoplatelets are contracted due to high compressive stress exerted by surface passivating amine ligands. The sharp feature of the (0002) plane indeed corroborates the lateral long-axis direction. These structural characterizations enable the schematic of a single nanoplatelet to be determined, as schematically illustrated in Figure 2f. A nanoplatelet consists of three pairs of parallel facets. The (1 1 20) is the dominant basal facet and it intersects (2Ϊ Ϊ0) facet with an angle of 30 degree. The (0002) facet is perpendicular to the long axis of the lateral dimension of the nanoplatelets and is normal to both (1 1 20) and (2 Ϊ Ϊ0) facets.

To elucidate the growth mechanism of nanoplatelet, electron microscope measurements were combined with absorption spectroscopy to characterize the intermediate products. The proposed the growth mechanism is shown schematically in Figure 3a: starting with bundled wires that fused in a second step into belts, which break into fragmented sheets and then undergo reconstruction to form slightly thicker sheets, and finally the proceeding sheets assemble through oriented attachment into nanoplatelets with uniform thickness. In the following section, ZnSe was used as an illustrative example of this process to substantiate the proposed mechanism.

As the reaction evolved for 2 min. at 150 °C, bundled nanowires were obtained with a length scale of a few micrometres (Figures 3b, 3f and 21) and each bundle consists of two to six individual nanowires. The absorption spectrum (1 in Figure 3k) manifests a bandgap at 323 nm, significantly blues-shifted from the bulk ZnSe bandgap of 459 nm due to the quantum confinement. HRTEM measurement on this aliquot shows that the width of each individual nanowires is uniform, which is ~0.99 nm (Inset of Figure 21 a). As the reaction proceeded for 4 min. at 150 °C, the absorption spectrum (2 in Figure 3k) shows an additional weak absorption peak at 348 nm and the dominant absorption features are similar to those of the preceding bundled nanowires, although the first exciton peak at 323nm (2 in Figure 3k) has a two-nanometre redshift and slightly sharpens.

TEM images of this aliquot (Figures 3c and 22a) show the product contains both bundled structures and small sheets, with the former of being the dominant phase. TEM measurement (Figures 3c and 22a) also shows that the bundled structures maintain the features of the preceding bundled nanowires. This confirms that the nanobelts are formed from the preceding bundled wires by fusing and stacking while maintaining their width, in consistence with the sharp peaks at 323 nm in the absorption spectrum (2 in Figure 3k). HAADF-STEM images (Figures 3g and 22b) show that at this stage the nanobelts already fragmented along their longitudinal direction although they still maintain the features of belts. The additional weak absorption peak at 348 nm (2 in Figure 3k) can be attributed to the presence of the sheets with small lateral dimensions and the 25 nanometres redshift of this absorption exciton peak also indicate the thickness of the sheets are larger than that of the preceding fragmented belts.

As the reaction evolved for 30 min. at 150 °C, the fragmented belts completely converted into nanosheets (Figure 3d and 3h). The absorption spectrum (3 in Figure 3k) presents two distinct sharp peaks at 348 nm and 331 nm, which correspond to the electron/heavy-hole and the electron/light-hole transitions. Further reaction at this temperature for two hours produces nearly monodisperse nanoplatelets (Figures 3e and 3i). The corresponding absorption spectrum (4 in Figure 3k) is identical to that of the preceding nanosheets (3 in Figure 3k), indicating the thickness of the nanosheets is exactly same as that of the final nanoplatelets. It is worth noting that some of the nanoplatelets show hollow and patchy features, as highlighted by white dashed circles in Figures 3i and 10. Combining the observation of the hollow and patchy features of nanoplatelets with the absorption spectroscopy data, it is concluded that ZnSe nanoplatelets are formed via an oriented mechanism from the preceding nanosheets. The changes of the dimensions of intermediates and nanoplatelets are depicted in Figure 3j and the absorption onsets in the absorption spectra (Figure 3k) are consistent with the increase of the width of the nanoparticles. Additional annealing at 150 °C for times ranging from 2 hours to 28 hours did not further change the dimensions of the ZnSe nanoplatelets. The growth mechanism studies on the wurtzite hexagonal ZnS nanoplatelets (Figure 23) show similar results, which corroborates the generality of this growth mechanism.

Computational Details

In order to understand the growth mechanism of ZnSe nanoplatelets, it is essential to detail the understanding of the growth morphology, which can be determined by both the surface energy and the growth kinetics. Here, these two aspects were studied to reveal the growth morphology mechanism of our ZnSe nanoplatelets with 1 4nm thickness by employed the first principles method based on the density functional theory (DFT) and the first principles pseudo potential method.

The DFT calculations were carried out by the Vienna Ab Initio Simulation Package (VASP) code [Kresse, G. & Furthmuller, J. Phys. Rev. B 54, 1 1169 (1996); Kresse, G.

& Furthmuller. Comput. Mater. Sci. 6, 15-50 (1996)] with the projector-augmented wave (PAW) method using the Perdew-Burke-Ernzerhof (PBE) exchange-correlation functional [Blochl, P. E. Phys. Rev. B 50, 17953-17979. (1994); Perdew, J. P. et al. Phys. Rev. Lett., 77, 3865 (1996)]. The plane wave cut-off was set as 500 eV, and the total energy convergence at 10 ^-6 eV for the self-consistent iterations. The Gaussian smearing method with s=0.05 eV was considered for Brillouin-zone integration and the geometry optimizations were stopped until the forces on the atoms were < 0.01 eV/A.

Figure 24 presents the ZnSe nanoplatelets mainly growing direction.

Perpendicular to the surface of the paper is the [0002] direction with paralleling to the c-axis, which corresponding to the three index [002] direction. The dominate surface is (1 1 20) with [1 1 20] direction opposite to the a3-axis, which corresponding to the three index (1 10) surface. The direction that intersects a1 -axis with an angle of 30 degree is [1010] direction. Along this direction, there are two different surface structures of (10Ϊ0)A and (10Ϊ0)B which corresponding to the three index (100)A and (100)B surface. Another growth direction with perpendicular to the [1 1 2 0] direction is [1 100], which corresponding to the three index [1 Ϊ0] direction. Along this direction, there are two different surface structures of (1 Ϊ00)A and (Ϊ 100)B. (1 Ϊ00)A is equivalent to (10Ϊ0)A corresponding to the three index (100)A surface. Whereas, the (1 Ϊ00)B surface structure is equivalent to (10Ϊ0)B with corresponding to the three index (100)B surface.

Surface energy

The finite-sized slab technique was used to model the surface. The slab models were built from the hexagonal ZnSe crystal structure determined by Korneev a

[Kristallografiya 6, 630-631 (1961)]. The lattice parameters of ZnSe bulk, optimised with a 12x12x6 k-point grid, were a=b=3.996A, c=6.626 A, which is in good agreement with experimental values. According to the exposed surface of ZnSe Nanoplatelets, five different surfaces were studied, they are non-polar surfaces of (1 1 20),

(1010)A/(1 Ϊ00)A and (10Ϊ0)B/(1 Ϊ00)B, and two polar surfaces of Zn-terminated (0002) and Se-terminated (0002 ), as shown in Figure 25. All these slabs models in periodical super cell were separated by a vacuum region of 20 A thick. Each slab of nonpolar surface models contain four zinc and four oxygen atoms within each Zn-O layer in a repeated slab configuration; while, for the polar surface models, each one contains eight zinc or eight selenium atoms within each Zn-O layer in the repeated slab configuration. Reasonable slab models were constructed by testing the slab thickness as illustrated in Table 3. The surface free energy of the particular slab with repeated geometry is given by the equation:

where n _t and m _i are the number of atoms and the chemical potential of the / th constituent of the slab. A is the surface area. E _si a _b is the total energy of the particular slab. The factor of 2 accounts for two equivalent surfaces in the particular slab. The surface energies of non-polar surface were found to converge with slab thickness and were converged to <0.001 eV/A ² with respect to eight layer slab thickness. We therefore using eight thick ZnSe slabs for studying the surface energy of (1 1 20) surface and (1010)/( 1 100) species surface with 3x6x1 and 6x3x1 Monkhorst-Pack k - point grid, respectively.

The computational treatment of polar surface requires special care, in particular to avoid the build-up of an overall artificial dipole field. Here, adopt symmetric structures of nine layer slab model were used for the Zn-terminated and Se-terminated 0002 surface with as shown in Figure 25 [Liu P.-L. & Siao, Y.-J. Scr. Mater. 64, 483-485, (201 1)]. The surface free energy equation is a thermodynamic function of the chemical potential of the constituent in the slab [Franchini, C. et al. Phys. Rev. B 73, 155402 (2006)]. Therefore, the surface energy is governed by the chemical potentials of the constituents of the slab, i.e. and m^. Invoking equilibrium of ZnSe bulk, m _£? ,

+ ii _St . = £¾/, the dependence can be further simplified by eliminating pse, leading to a dependence on only. The ranges of _z to the $t§*, _¾ + AH _/ [ZnSe] < ie

AH/[ZnSe] < £ 0, were restricted according to the thermodynamically allowed ranges in surface. These ranges are determined by the assumed constraints, , which mean the Zn and Se do not crystallize in the surface, and moreover, the formation enthalpy of bulk ZnO given

b _{zc b} · We have calculated surface energies as linear functions of ¾ _¾ to -2.360 eV < Dm _Zΐ„ £ 0 as shown in Figure 4a. The chemical potential Am _å = 0, i.e. , represents the Zn-rich condition; Am _Z! , = -2.360 eV the se-rich condition.

Table 3. The surface energy of (1 1 20), (1010)A/(1 100)A and (1010)B/(1 Ϊ00)B facets as a function of thickness.

6 layers 7 layers 8 layers 10 layers

(1 1 20) 0.0158 0.0153 0.0153

(10 ϊ 0)A/(1 Ϊ00)A 0.0166 - 0.0161 0.0161

( 10 Ϊ 0) B/( 1 Ϊ00)B 0.0564 - 0.0557 0.0557

The (1 1 20) is most stable surface among all the studied surfaces of ZnSe nanoplatelets. So there would be a very strong (1 1 20) preferred-orientation in growth of ZnSe nanoplatelets. (1010)A/(1 100)A are also very stable with only a little larger surface energy above (1 1 20), but (1010)B/(1 Ϊ00)B have great larger surface energy than other non-polar surfaces. So the (10 Ϊ 0)B/(1 Ϊ 00)B might play a role as driving force to make the growing energetically along [10 ϊ 0]/[1 Ϊ 00] direction, which preferred- terminate at (1010)A/(1 100)A surface. Whereas growing in [1010] and [1 Ϊ00] direction might combined their action to the fast growth of [1 Ϊ 00] direction. Moreover, two polar surface of (0002) and (0002 ) both have very larger surface energy than any other non-polar surfaces over almost all the thermodynamically allowed ranges, only excluding the surface energy of (0002 ) as Am _z below -2.1 eV in Se rich condition. These would result in very fast grow along [0002] (c-axis). Further, the (0002 ) is more stable than (0002) over a wide range of chemical potential, which indicates that (0002 ) is the preferred-termination of the [0002] direction. For a view, the surface energy calculations show the (1 1 20) surface is the facet-domination. And the ZnSe growing would be fast along [0002 ] (c-axis) and [1 Ϊ 00], which would preferred-terminate at (0002 ) and (1010)A/(1 Ϊ00)A surface, respectively. These preferred growing processes might finally result in the forming of ZnSe nanoplatelets.

ZnSe monomer and alkylamines binding energy

According to the experiment, the growing environment of ZnSe nanostructure contain the Zn cation, Se anion, and alkylamine ligand, which might directly influent the growing behaviour. So it is essential to accurately investigate the binding behaviour of alkylamines ligand and ZnSe monomer on the exposed surfaces of ZnSe

nanostructure. Here we also use the slab model with eight atomic layer separated by a vacuum space 20 A to study the features of molecule binding to the surfaces. The surface sizes are the same as the models of studying the surface energy above with similar k-point meshes. In order to suppress any long-ranged interference between the Zn and Se-terminated ends of the slab, the dangling bonds of the bottom atoms of ZnSe (0002) and (0002) slab models were saturated by adding one monolayer of pseudo hydrogen atoms with nuclear charge Z = 1/2|e| and Z = 3/2|e|, respectively.

The binding energy is calculated as E _bi nding= Etot-E _Siab -Eamine for alkylamines adsorption and E _b inding= (Et ₀ rEsiab)/n-E _Z n-Ese for ZnSe monomer adsorption. Where E _to t and E _siab , are the total energy of the adsorption system, the energy of slab model without adsorption, respectively; Eamine, E _zn and E _se are the energy of the alkylamines, bulk Zn, and bulk Se, respectively; n is number of ZnSe monomers. To simplify calculations, methylamine served as a model for alkylamines. For the adsorption of methylamine, four-symmetry adsorption sites that are top(Zn)-site, top(Se)-site, bridge- site, and hep-site were simplified. For the ZnSe monomer, the ZnSe monomer adsorption on crystal space lattice site was considered.

Table 4 shows the binding energies of the methylamine and ZnSe monomer adsorption states on different sites of every surfaces, and their adsorption optimized structures were all presented on Figure 26. The most stable methylamine adsorption for each surfaces were listed as follow: (1 1 20) hep-site with -0.9317 eV,

(1010)A/(1 Ϊ00)A hep-site with -0.8876 eV, (10Ϊ0)B/(1 Ϊ00)B with -3.7583 eV, (0002) with hep-site with -2.9506 eV, (0002) top-site with -5.1486 eV and (0002) hep-site with -1 .4514 eV. By contrast, ZnSe monomer adsorption on the crystal space lattice site of each surface was (1 1 20) hep-site with -1 .8565 eV, (10 Ϊ 0)A/(1 Ϊ00)A with -1.7029 eV, (10Ϊ0)B/(1 Ϊ00)B with -3.7583 eV, (0002) with -5.6899 eV and (0002) with -3.5473 eV, each ZnSe monomer adsorption states has much lower binding energy than that of methylamine adsorption. These results illuminate that all the surfaces prefer to binding with ZnSe monomer prior to the alkylamines. So the ZnSe growth is unlikely inhibited by binding with alkylamines, and the epitaxial growth of all the surface would be promoted by the strong binding with ZnSe monomer. Especially, the surfaces of (1010)B/(1 100)B, (0002) and (0002), who also have large surface energy, show strong binding energies with ZnSe monomer, which would contribute to the fast growing along c-axis and [1 100] direction. Table 4. The formation energy of diffusion and stacking process and fuse process along [11 20] direction and [1 Ϊ 00] direction, respectively.

Diffusion process

7 ^th -layer 8 ^th -layer 9 ^th -layer Fuse process

^{[10 l 0]} 0.0357 -0.2489 0.0655 -0.5979

[1210] .2 7828 0.8364 -2.5247 -7.0577 (Horizontal) -6.9369 (ladder )

ZnSe monomer and alkylamines binding energy in nanowire model 5 As diameter of initially formed zinc amine lamellar structure is determined by the template, the ZnSe nanowire was used as a model with a diameter of about 0.9 nm. Figure 27 shows the nanowire model structures, which contain two equivalent structures. For the [1 1 20] direction, there are six atomic layers with thickness of 0.99 nm, while for the [1 Ϊ00] direction, the distance between two terminated (1 Ϊ00)B 10 surfaces is 0.92 nm as shown on Figure 27. It is to be noted that in case of nanowire it is periodic along the c-axis, and sufficient vacuum of 20 A has been kept along [11 20] and [1 Ϊ00] direction. And 8x1 x1 Monkhorst-Pack /(-point grid has been taken for the total energy calculations. These ZnSe nanowire models are built after serious consideration, because five or seven atomic layers along [1 1 20] direction have 15 thickness of 0.79 nm and 1.19 nm; while the distance between (1 Ϊ00)B and (1 Ϊ00)A or two (Ϊ 100)A along [1 Ϊ00] direction are 1.04 nm and 1.17nm. So these sizes are far away from the reasonable diameter as the measured value of 0.9nm in the experiment. The nanowire models above was used as the significant starting point, in particular study the growth kinetics of ZnSe nanoplatelets in next section.

2 o From Figure 28, for (11 20) surface of nanowire, all the average binding energy with different binding number of ZnO monomer are below -2.3 eV. For (1 100)B surface, the average binding energy has a slight increasing trend, but all the average binding energy with different binding number of ZnO monomer are below -3.0 eV.

Thus, from ZnO monomer binding behaviour, the nucleation could probably proceed on

25 both of (11 20) nor (1 100)B surfaces. In addition, the octylamine and ZnSe monomer binding behaviour with the nanowire model was also investigated, which is similar to the behaviour of the methylamine with slab models. The octylamine show

physisorption state neither on (1 1 20) nor (1 Ϊ00)B surface with binding energy of - 0.842 eV and -0.984 eV, respectively. Overall, the ZnSe monomer show much stronger

30 binding behaviour with both of (1 1 20) and (1 Ϊ00)B surface than octylamine for

nanowire. So, for nanowire preferred binding with ZnSe monomer, this behaviour could promote the epitaxial growing along of [11 20] and [1 Ϊ00] direction. ZnSe nanoplatelets growth kinetics

In this section the atomic scale step structures on ZnSe nanoplatelets growth was studied. The step structures and corresponding formation energies determine the differences of surface morphologies and growth rate between the two different surfaces. Through researching on the energetic and structural properties of steps structure, their influences on the formation of ZnSe nanoplatelets were determined.

The atomic scale step model starts with the nanowire with a diameter of about 0.9 nm as shown in Figure 27.

It has been confirmed that growth along [0002] c-axis is energetic and fast due to the great large surface energy of (002) polar surface as descripted above. The final morphology of ZnSe nanoplatelets might be mainly dominated by the unclearly growth behaviour along the direction of [1 1 20], [1010] and [1 Ϊ 00], so the study mainly focused on growth behaviour along these directions. There might be two possible growing ways along these two directions (see Figure 4c and 4d in main article). One is that ZnSe monomers diffused to the nanowire surface, reconfiguration and form the new surface layer, thereafter more ZnSe monomers would repeat the process and result in the stacking the steps of layer by layer to increase the thickness of that direction. The other might be through two or more nanowires oriented to fuse each other, then reconstructed to promote the growing along that direction. Both of these two possible ways were studied to determine the growth on the each other [1 1 20] and [1 Ϊ00] direction, respectively, and it was determined that the reason of growth morphology of 1 4nm ZnSe nanoplatelets with (1 1 20) as the dominate surface.

The formation energy can be defined and calculated as:

Here, E _to tai and E _te rrace are the DFT total energy of the as-simulated step model and the pure terrace parts in the same model, and n _{z n} , _¾ , n _Se and are the number of Zn atoms and Se atoms in the system and their chemical potential, respectively. It was imposed that the terrace parts of ZnSe nanocrystal always has the thermodynamic equilibrium conditions as E _te rerac/n _mo nomer, here n _mo nomer is the number of ZnSe monomers in terrace parts of ZnSe nanocrystal.

Then, the formation energy of new ZnSe nanostructures was calculated which formed by stacking layer by layer along [1 1 20] and [1 100], respectively (see Table 5). For the [1 1 20] direction, the formation energy of step ZnSe nanostructure after stacking the first, second and third atomic layer are 0.0359 eV, -0.2489 eV and 0.0654 eV with corresponding to the thickness of 1 19nm, 1 39nm and 1 59nm along

[1 1 20] direction, respectively. The small positive formation energy value of 0.0359 eV as growing the first additional layer along [1 1 20] direction implies a little energy barrier during this nucleation process with corresponding to 415 K to form a new atomic layer and increase thickness from 0.99 nm to 1 .19 nm. It would be a spontaneous nucleation to form second additional layer and result in thickness of 1 .39 nm along [1 1 20] direction due to the negative formation energy of -0.2489 eV. But sequentially growing to increase the thickness to 1 6nm at [1 1 20] direction would be very difficult as the formation energy add to 0.0654 eV. It means the environment should provide more than 760 K energy to overcome the energy barrier of 0.065 eV, which is beyond the experimental conditions. So, the final thickness along [1 1 20] direction of ZnSe nanostructure might be 1 39nm corresponds well to the experimental result of 1 .4 nm. For the [1 Ϊ 00] direction, the formation energy of ZnSe nanostructures after stacking first, second and third atomic layer are -2.7826 eV, 0.8365 eV and -2.5247 eV, respectively. The addition of first and third surface layer correspond to the (1 100)A surface structure, have very low formation energy. So it would be very fast growth of (1 Ϊ00)A surface. However, the formation of second surface layer with corresponding to (1 Ϊ00)B surface structure is positive of 0.8365 eV, which implied the difficulty nucleation on (1 Ϊ00)B surface. These results agree with the surface energy study above. Whereas (1010)A/(1 Ϊ00)A has very low surface energy, these surfaces would be fast formed and tend to be the terminal surfaces of [1010] and [1 100] directions; nevertheless, (1010)B/(1 Ϊ00)B has very high surface energy, so nucleation would be difficult to form (1010)B/(1 100)B surfaces, and these surfaces tend to be disappear after the whole growing process. Further, the formation energy of new ZnSe nanostructure coupled by two nanowires oriented to fuse along [1 1 20] and [1 Ϊ00], respectively, was calculated. The results found to be -0.5975 eV, -6.937 eV and - 7.058 eV for the [1 1 20] oriented direction, ladder type [Ϊ 100] oriented direction and horizontal type [1 Ϊ00] oriented direction, respectively (the fuse configuration see Figure 4c). So the nanowires oriented fuse along [1 Ϊ00] direction have much lower formation energy than that of [1 1 20] direction, which indicates the nanowires would preferred to oriented fuse and fast growing along [1 Ϊ00] direction other than [1 1 20] direction. Moreover, the nanowires might prefer taking the horizontal type to fuse together along [1 Ϊ00] direction, due to the lower formation energy of horizontal type than that of ladder type. In a view of two growing ways at [1 1 20] and [1 100] direction, nanowires oriented to fuse as horizontal type along [1 100] direction have the great lower formation energy than any other growing process. So, nanowires horizontal type fuse along [1 Ϊ 00] might be the dominated growth process, which result in many small lateral size of nanobelts. These small lateral sizes of nanobelts would continuously orient with each other to attach to each other to form larger lateral size of nanoplatelet. Within this process, ZnSe monomers would diffuse and nucleate on the (11 20) surface, which determines the final 1 .39 nm thickness along [1 1 20] direction of ZnSe nanoplatelets. The large lateral size might be too heavy to move to each other to continuously the oriented attachment, the ZnSe monomers would diffuse fast and nuclei constructed on exposed (1 Ϊ00)B and (1010)B, and form the (1 Ϊ00)A and (10Ϊ0)A terminated to finish the formation of the ZnSe nanoplatelets.

Table 5. The binding energy of alkylamines ligand and ZnSe monomer adsorption on the different sites of exposed surfaces of ZnSe nanostructure. Top-Zn Top-Se Bridge Hep Monomer

(11 20) -0.9273 0.0435 -0.0323 -0.9317 -1.8565

(1010)A/(1 Ϊ00)A .0.7396 -0.0379 -0.8577 -0.8876 -1.7029

(10 ϊ 0) B/(1 Ϊ00)B -1.1160 0.0064 -1.1160 -2.9506 -3.7583

(0002) -5.1486 -4.6258 -3.8125 -5.6899

(0002 ) -1.4489 -1.4494 -1.4514 -3.5373

To further establish the mechanism proposed above and to study the surface energy and growth kinetics of nanoplatelets, in particular to elucidate why

nanoplatelets form uniform thickness of 1 .39 nm, the first principles method based on the density functional theory (DFT) and the first principles pseudo potential method were employed to solve numerically for nanoplatelets and the preceding nanoparticles (Figure 24).

Firstly, the finite-sized slab technique (Figure 25) was used to study the exposed facets of nanoplatelets (Figure 4a, Table 4). The (1 1 20) is the most stable facet thus this facet is expected to be the dominant facets that determines the final morphology of the products. (1010)A/(1 100)A are also very stable with the surface energy being slightly larger than that of (1 1 20) (Inset of Fig. 4a), but (1010)B/(1 Ϊ00)B have much larger surface energy than other non-polar surfaces. Therefore, the high surface energy of (10Ϊ0)B/(1 Ϊ00)B would play a role as driving force to promote the growth along [10Ϊ 0]/[ 1 Ϊ 00] direction with (1010)A/(1 Ϊ00)A as terminated face. Two polar facets of (0002) and (0002) both have very larger surface energy than any other non-polar surfaces over almost all the thermodynamically allowed ranges (Fig.4a) and this will lead to fast growth along [0002] (c-axis). Moreover, (0002) is more stable than (0002) in a wide range of chemical potentials, indicating that (0002) is the terminated facet of the [0002] direction. The calculation results on the surface energy are in good agreement with the developed facets of nanoplatelets depicted in Figure 2g.

Next, to investigate if the nucleation and growth will proceed reasonably or be inhibited by the ligand binding the binding behaviours of zinc monomer and alkylamines ligand on the exposed facets both of slab models and nanowire model were studied (Table 4, Figures 36 and 28). For either slab models or nanowire model, the ZnSe monomer always show much stronger binding ability on any surfaces of nanoplatelets than that of alkylamines ligand. So, the ZnSe growth is unlikely inhibited by binding with alkylamines, and the epitaxial growth on each surface would be promoted by the strong binding with ZnSe monomer. Moreover, the surfaces of

(10l0)B/(1 Ϊ 00)B, (0002) and (0002), which have large surface energy (Figure 4a), also show strong binding energies with ZnSe monomer. This effect would promote the epitaxial growth on (1010)B/(1 Ϊ00)B and fast growth along (0002) direction (c-axis). As the dimension of the nanowires is determined by the preformed structure of zinc precursor and alkylamine ligand, the nanowire model with periodic repetition along the c-axis and a diameter of ~0.99 nm was used as the starting point to investigate the growth of the nanowires along [1 1 20], [1010] and [1 100] directions from an atomic-scale perspective (Figure 4b and 27). Two possible growth pathways were proposed. One might be the orientation to fuse of two or more nanowires followed, then the fused nanowires undergo reconstruction to promote the growth along the fusing direction (Figure 4c1 and 4c2). In this case, [1 1 20] and [1 Ϊ00] directions were the main orientation fuse directions for nanowire model, as the geometric accumulation of nanowire model could be continuous along these two directions. The other pathway is that ZnSe monomers diffuse onto the surface of nanowire, reconstruct and form the new surface layer, thereafter more ZnSe monomers would repeat this process, resulting the increase of the thickness of that direction by layer by layer stacking (Figure 4d1 and 4d2). From ZnO monomer binding behaviour, the nucleation could probably proceed on both of (1 1 2 0) nor (1 Ϊ00)B surfaces of nanowire (Figure 28).

In view of these two growth ways, horizontal type orientated fusion along [1 Ϊ 00] direction (Fig. d and c2) might be the dominated growth process as it has much lower formation energy than any other fusing process (Figure 4c3 and Tables 5). This oriented fusion may not only contribute to the fast growth along [1 Ϊ00] direction, but also produce the dominated (1 1 20) facet of ZnSe nanoplatelets. This growth process can be continuously repeated and finally result in the formation of small lateral size ZnSe nanobelt structure with thickness of 0.99 nm along [1 1 20] direction.

As the nanowires may be depleted by the fuse process, in the next step, ZnSe monomer diffusion and reconstruction become the dominant process that leads to the form additional atomic layers on the surface of nanoplatelets. The nucleation and growth along the [1 1 2 0] direction should overcome a small energy barrier of 0.0359 eV, which is equivalent to 415 K (141 .85 °C), to form a new atomic layer and increase thickness of nanobelt from 0.99 nm to 1 .19 nm (Figure 4d1 and 4e). This is energy barrier can be easily bridged as our experiment is conducted at 150 °C. After the first additional layer being deposited on (1 1 20) surface, it will spontaneously form the second additional layer and increase the thickness of nanobelt to 1 .39 nm due to its negative formation energy of -0.2489 eV (Figure 4e, Table 5). However, after the thickness of the nanobelt increases to 1 .39 nm, to deposit any further atom layer along [1 1 20] direction is inaccessible as the formation energy to form the third additional layer vastly increases to 0.0654 eV (760 K) and it is far beyond the experimental conditions. So the final thickness along [1 1 20] direction of ZnSe nanostructures is 1 .39 nm, being in consistent with our experimental result.

Next, 1 .39 nm thick nanobelt with small lateral size will continuously be oriented to attach along the [1 Ϊ00] direction to form nanoplatelets with large lateral size. The exposed (1 100)B and (10Ϊ0)B will grow fast due to monomers diffusion and reconstruction, then form new atomic layers with the (1 Ϊ00)A and (10Ϊ0)A structures due to the large surface energy of (1 Ϊ 00)B and (10Ϊ0)B (Fig. 4a). Furthermore, the behaviour of ZnSe monomers diffusion and reconfiguration along [1 100] and [1010] also demonstrate the formation of (1 Ϊ00)A and (10Ϊ0)A is much easier than (1 Ϊ00)B and (10Ϊ0)B (Figure 4d2, 4e). So the new surface of (1 Ϊ00)A and (10Ϊ0)A would be the terminated surface in their corresponding directions as further growing another atom layer on these surfaces to form (1 Ϊ 00)B and (10Ϊ0)B was very difficult due to the large formation energy of 0.836 eV (Figure 4f and 29). All these growth behaviours of ZnSe nanoparticles above would finally result in ZnSe nanoplatelets with uniform 1 .39nm thickness along [1 1 20] direction and well-developed facets as schematically shown in Figure 2g.

Conclusion

Monodisperse ZnSe nanoplatelets were obtained in the syntheses in which varieties of amines (either single amine or a co-solvent containing a short carbon chain amine and oleyalmine) (Figure 30) and zinc precursors (Figure 31) were used. These experiments clearly demonstrate the robustness of this synthetic route.

Furthermore, this synthetic route has been further extended to synthesize a variety of nanoplatelets with diverse compositions such as alloyed zinc chalcogenide (ZnSe _x Si- _x , ZnTe _x Sei_ _x , x=0-1 ), Sb ₂ S ₃ , CU2S and B12S3 nanoplatelets (Figure 5), demonstrating the generality of this synthetic approach. It was also shown that ZnS nanoplatelets can convert into CdS, Cu ₂ S and MoS ₂ via a cation exchange reaction (Figure 5), producing libraries of nanoplatelets with uniform thickness. All these enable a method to precisely tune the band gap of nanoplatelets, as illustrated in Figure 5k.

It is of significant importance to further tune the thickness of the nanoplatelets as the quantum confinement effect of 2-D semiconductor nanomaterials is mainly determined by the thickness of this dimension. The temperature of the reaction system may be a critical factor that may be utilized to further tailor the thickness of nanoplatelets. When the temperature of the reaction was increased from 170 °C to 230 °C, it was observed that the morphology transition of ZnSe from nanoplatelets to nanowires accompanied by the change of the smallest dimension of nanoparticles (Figure 32). This suggests that thickness control of nanoplatelets is not accessible by changing the reaction temperature within the scope of the synthetic conditions, which is in good agreement with the results of the formation energy calculation regarding depositing an additional layer along the [1 1 2 0] direction on the 1 .39 nm uniform thickness nanoplatelets.

ZnS and ZnSe nanoplatelet-based UV detectors

Fabrication of photodetectors

The ZnSe nanoplatelet powders were dispersed in toluene (10 mg mL· ¹ ) with sonication. Interdigitated electrode arrays on quartz substrates were cleaned subsequently with detergent water, acetone and isopropanol for 30 minutes with sonication, then blow dried with nitrogen and treated with oxygen plasma. The electrodes have a length of 1 mm, an electrode width of approximately 20 pm and a spacing between adjacent electrodes of approximately 10 pm. 10 pL ZnSe/toluene was dropped on the cleaned interdigitated electrode and the solvent was evaporated at room temperature. The sample was then baked at 300 °C in nitrogen for 30 minutes. Photodetector response

Photo responses to 365 nm and 254 nm UV light were studied for ZnSe nanoplatelets-based photodetectors. Current-Voltage characteristics were obtained by sweeping the voltage from -100 V to 100 V at 1 V/step with and without UV light irradiation. As shown in Figure 35B, the device has obvious photo-induced current for both wavelengths. Figure 35C and 35D show the transient responses of the photodetector to 365 nm and 254 nm UV light, respectively. In both cases, the device currents exhibit a step rise and fall with a response time of about 0.3 seconds (Figure 36) upon turning on and off the UV light. The photo-induced current at 100 V is about 7 nA excited by the 365 nm UV light and about 1.5 nA by the 254 nm UV light, suggesting that the photodetector has a spectrum responsivity of about 3.5 mA W ^-1 at 365 nm and 0.6 mA W ¹ at 254 nm, respectively. In comparison, commercial gallium phosphide (GaP)-based UV detectors have a responsivity of about 90 mA W ^-1 at 365 nm and 30 mA W ¹ at 254 nm, respectively. Nonetheless, the fabrication of the GaP UV detectors usually requires high temperature and high vacuum processes and expensive single crystalline epitaxial substrates.

2-D nanoplatelets have a giant oscillator strength transition, which will be manifested by a significantly short exciton radiative decay time. To this end, ligand exchange was performed using tributylphosphine (TBP) or trioctylphosphine (TOP) to increase the fluorescence quantum yield by passivating the surface traps of nanoplatelets. TBP was used to substitute the surface passivating ligand of alkylamine for ZnSe nanoplatelets whereas TOP was used for the case of ZnS. The diffraction peaks in the XRD patterns for both ZnSe and ZnS after the surface ligand has been changed into TBP or TOP do not show any shift with respect to those of the original nanoplatelets (Figure 40). The standard XRD patterns for both wurtzite ZnSe and ZnS were given for reference. TBP was used to substitute the surface passivating ligand of alkylamine for ZnSe and alloyed nanoplatelets whereas TOP was used ZnS nanoplatelets. The absorption spectra for all nanoplatelets retain the features of the original nanoplatelets after the surface ligand has been changed into TBP or TOP although a slight red-shift was observed (Figure 41).

Alloyed ZnSi- _x Se _x (x=0.25, 0.50) nanoplatelets were synthesized to further tailor the band gap of nanoplatelets (Figure 37A and 37B), which enabled tuning the fluorescent emission between 296 nm and 349 nm (Figure 37 C and 37D). The absorption spectra for ZnSe and ZnSo ₅₀ Seo ₅₀ nanoplatelets show distinct light hole electron (Ih-e) and heavy hole-electron (hh-e) transitions, typically of the electronic structure of quantum wells (Figure 37C). However, these transitions are not resolved for ZnS and ZnSo ₇₅ Seo ₂₅ nanoplatelets, possibly because their small spin-orbit splitting, since the spin-orbit coupling is found to decrease with decreasing the atomic number of the chalcogens of ZnX (X=Te, Se, S). This finding is in consistence with that for CdS nanoplatelets whose Ih-e and hl-e transitions are also not resolved because of the small spin-orbit splitting of the valance bands. The emission spectra of nanoplatelets show extremely narrow emission bands with full-width at half-maximum as sharp as 90 meV for ZnSe nanoplatelets (Figure 37D), suggesting atomically flat and extended surfaces of these quantum wells. The fluorescence emission band of ZnSe nanoplatelets continue to shift to lower energy as the temperature decreases from room temperature to 77 K (Figure 37E and Figure 38). The fluorescence lifetimes for zinc chalcogenide nanoplatelets are ~2 ns at room temperature (Figure 37F) and are slightly shorter at 77 K (Figure 39). Such lifetimes are more than two orders of magnitude faster than that for spherical ZnSe nanoparticles and are only about a half of that of CdSe nanoplatelets. The extremely short fluorescence lifetimes can be attributed to the giant oscillator strength transition of quantum wells with strong spatial confined 2-D excitons.

In the claims which follow and in the preceding description, except where the context requires otherwise due to express language or necessary implication, the word “comprise” or variations such as“comprises” or“comprising” is used in an inclusive sense, i.e. to specify the presence of the stated features but not to preclude the presence or addition of further features in various embodiments of the method and nanoplatelets.