FIELD OF THE INVENTION

The present invention is in the field of a full wafer transfer-free graphene layer, a method of making the graphene layer, and a product obtained by said method, such as sensors, such as gas sensors, MEMS, IC's, PV-cells, solar cells, mi croscale and nanoscale devices, and products having the small scale devices.

BACKGROUND OF THE INVENTION

Graphene is carbon comprising material. Its structure relates to one-atom-thick planar sheets of sp2-bonded carbon at oms that are crystallographically densely packed in a honeycomb crystal lattice. The crystalline or "flake" form of graphite consists of many graphene sheets stacked together.

It can be a basic building block for graphitic materials of all other dimensionalities. It can be wrapped up into fuller- ene, rolled into ID carbon nanotubes or stacked into 3D graph ite .

Graphene has attracted a lot of research interest because of its promising electronic applications related to its superior electron mobility, mechanical strength and thermal conductivity. It may have wide range of applications, for instance, field-effect transistors, photonic or optoelectronic device, sequencing DNA through nano-holes in graphene etc. Graphene macroscopic samples have unusual properties such as a bipolar- transistor effect, ballistic transport of charges, large quan tum oscillations, etc.

Various production methods of graphene are reported. Gra phene or ultra-thin graphitic layers can be epitaxial grown on various substrates. Graphene produced by exfoliation was a very expensive material. Since then, exfoliation procedures have been scaled up. It is noted that the price of epitaxial grown graphene on e.g. SiC is dominated by the substrate price. Graphene has been produced by transfer from various metals and alloys thereof, though graphene may be slightly rippled .

It remains however difficult to obtain high quality and clean graphene e.g. in a device. Compared to monolayer gra- phene, crystallized multilayer graphene has stronger mechani cal properties and higher conductivity. Such graphene is still transparent under the optical microscope and electron microscopy, and has high potential in the field of nano-imaging technology. It remains also difficult to transfer graphene such that the quality and integrity thereof are maintained at a high standard, which is crucial for the characteristics of graphene. The prior art graphene is not clean, has lots of contamination and cracks.

For graphene transfer from a metal surface various methods may be used. However transfer is complex and risky; a graphene layer is typically damaged, unless dedicated measures are taken. During the transfer, the graphene is prone to stretch ing by a rigid (plastic or glass) substrate that may be used for transfer, and the substrate always contains lots of exter nal contaminations. So, this method has various drawbacks: Introduction of external particles and contaminations; Genera tion of strong strain on the graphene, the graphene films may tear and form residual stress and cracks. The transferred gra phene contains lots of bubbles, which makes the quality of graphene at least one order of magnitude lower; the graphene obtained is difficult to handle. The graphene sometimes can be attached directly to the plastic or glass substrate. Then is it impossible to separate them anymore.

Various documents recite graphene synthesis.

Zhangcheng Li et al, in "Low-temperature growth of graphene by chemical vapour deposition using solid and liquid carbon sources", ACS Nano, 2011, 5, 3385-3390, use benzene as carbon source and a quartz tube hot wall furnace to synthesize gra phene at low temperature. Therein a copper foil (25 pm) was used as a synthesis substrate, providing only microscale graphene flakes at 500 °C. The method does not provide large scale graphene, nor high quality, nor graphene that can be harvested easily.

Y. Kim et al, in "Low-temperature synthesis of graphene on nickel foil by microwave plasma chemical vapour deposition", Appl. Phys. Lett., 2011, 98, 263106, use a cold wall type mi crowave plasma chemical vapour deposition to synthesize centi metre scale graphene at low temperature. Therein as feed stock methane and hydrogen are used, as substrate material a poly crystalline nickel foil (50 pm) is used, within a temperature range from 450 °C to 750 °C. Thereto a microwave with a power of as much as 1400 W was used to generate a plasma, which at tributes to a high energy consumption.

Gopichand Nadamuri et al, in "Remote plasma assisted growth of graphene films", Appl. Phys. Lett., 2010, 96, 154101 use a horizontal oven with a quartz tube {3 cm diameter} which is surrounded by a copper RF-coil on one side. Graphene is grown on a 300 nm nickel film, nickel foil, and nickel single crys tal. The RF-coil with 250 W power can induce a plasma 25 cm from the substrate. Graphene was synthesized with methane and hydrogen as feed stock at 650 - 700 °C with a base pressure of 1 Torr.

Jaeho Kim et al, in "Low-temperature synthesis of large- area graphene-based transparent conductive films using surface wave plasma chemical vapour deposition", Appl. Phys. Lett., 2011, 98, 091502 employ a 3 - 4.5 kW surface wave plasma chem ical vapour deposition to grow graphene on a 23 cm x 22 cm copper foil (30 pm) and an aluminium foil (12 pm) at 300 - 400 °C with a gas pressure of 3 - 5 Pa. Methane, argon and hydro gen were used as feed stock. A sheet resistance ranged from 2.2 kQ to 45 kQ per square with a transmittance of 78% to 94%.

Qingkai Yu et al, in "Control and characterization of indi vidual grains and grain boundaries in graphene grown by chemi cal vapour deposition." Nature Materials, 2011, 10, pp 443- 449, used copper as substrate, and a quartz tube hot wall furnace to synthesis graphene. Single hexagonal graphene crystals were achieved with controllable patterning and high tempera ture growth under ambient pressure condition at 1050 °C. How ever, the graphene crystal size was limited to tens of micrometres due to a bad control of the process.

Xuesong Li et al, in "Large-Area Graphene Single Crystals Grown by Low-Pressure Chemical Vapour Deposition of Methane on Copper", J. Am. Chem. Soc., 2011, 133, pp 2816-2819, used a folded copper foil as substrate to grow graphene under very low pressure condition (with a methane flow rate of 0.5 seem corresponding to partial pressure of 8 mTorr) at 1035 °C with a quartz tube hot wall furnace. The graphene size up to 0.5 mm was achieved within 90 min. However, under the conditions used the results obtained are non-reproducible.

Hong Wang et al, in "Controllable Synthesis of Submillime ^¬ tre Single-Crystal Monolayer Graphene Domains on Copper Foils by Suppressing Nucleation", J. Am. Chem. Soc. , 2012, 134, pp 3627-3630, used a copper foil as substrate to grow sub-milli metre graphene under ambient conditions. A large amount of hy drogen gas (500 see ) was used to suppress the nucleation of graphene. However, the graphene obtained had a rectangular shape, and the quality of graphene was bad, it could be dam aged easily.

Takayuki Iwasaki et al, in "Long-Range Ordered Single- Crystal Graphene on High-Quality Hetero-epitaxial Ni Thin Films Grown on MgO(lll)", Nano Lett., 2011, 11, pp 79-84, used a Ni thin film to growth large single crystal graphene on MgO (111) single crystal substrate. This method however is unsuit able for mass production.

Libo Gao et al, in "Repeated growth and bubbling transfer of graphene with millimetre-size single-crystal grains using platinum", Nature Communications 2012, 3, 699, grow graphene on a platinum foil substrate in a quartz tube hot furnace at 1000 °C. This method however is unsuitable for mass produc tion.

It is noted that various methods relating to synthesizing other carbon comprising molecules, such as carbon nanotubes, are known. These methods typically are not applicable for ob taining graphene.

Further prior art documents relate to graphene formation. For instance Vollebregt in IEEE, MEMS 2016, Shanghai, China, 24-28 January 2016, p. 17-20, recite a transfer-free wafer scale CVD graphene fabrication process for MEMS/NEMS sensors. Therein phosphoric acid is used as an etchant, which etchant is better not used in view of health issues and quality of graphene obtained. The fabrication process results in sub-optimal graphene being formed; for instance for relatively small device the yield is between 89% and 97%; the resistivity is two order of magnitude higher than possible, which is at tributed to a high defect density; also the transmission is rather low (59.5% at 550 run). Vollebregt in Transducers 2017, Kaohsiung, Taiwan, June 18-22, 2017, p. 770-773, recites suspended graphene beams with tuneable gaps for pressure sensing, wherein said beams relate to small, typically micrometer scale, structures. Also this fabrication process results in sub-optimal graphene being formed; for instance the structures obtained reflect a yield of about 98%; the sheet resistance is relatively high, namely about 4 kQ; also the defect density is too high, which follows from the Raman data.

A drawback of prior art methods is that the quality of the graphene is not very good, e.g. it may contain many dislocations. Further it is difficult to grow a large area of gra phene layers, especially of good quality. Typically when ob taining graphene after growth thereof it is cumbersome to separate graphene, such as by removing a supporting layer. It is noted that various techniques, e.g. PECVD, result in poor quality graphene. It is also a drawback that prior art systems are not very costs effective, e.g. as synthesis consumes relatively large amounts of energy, are performed at relative high temperatures (1000 °C or higher), etc. As a consequence also characteristics of a graphene layer are not very good, e.g. in terms of being impermeable to gas and liquid, in terms of ho mogeneity, in terms of conductivity, etc.

The present invention therefore relates to an improved method of obtaining a graphene layer on a silicon wafer, which solve one or more of the above problems and drawbacks of the prior art, providing reliable results, without jeopardizing functionality and advantages.

SUMMARY OF THE INVENTION

The present invention relates to a method of providing a graphene layer on a substrate, comprising providing a substrate, such as a silicon wafer (10) , wherein the graphene layer covers the substrate for 50-100%, such as covering a full substrate, which may be any wafer of any size, such as commercially available silicon wafers. The method is applica ble to any size of wafer, e.g. from 4-28 inch, or larger, and in principle to any flat surface as well. On the silicon wafer a diffusion barrier layer, such as a dielectric layer (11) , is provided. The diffusion barrier may be deposited using chemi- cal or physical vapour deposition. The term "on" may be re flecting that subsequent layers are in direct contact with one and another, or that further intermediate layers are present. Further on the dielectric layer a sacrificial transition metal layer (12) is provided and on the metal layer a graphene layer (13) is provided. If desired, the metal layer, e.g. Mo layer, or Si wafer, can be pre-patterned using lithography and etching. The graphene may be deposited on a Mo-based layer using catalytic chemical vapour deposition (CVD) . It has been found that is important that prior to providing the graphene layer a surface roughness of the metal layer surface is reduced.

Therewith a much better graphene quality is obtained, such as in terms of defect density thereof, a yield thereof, optical properties, and electrical properties. After the graphene dep osition on the metal layer, it is possible to perform post processing on the wafer, for instance, to integrate electrical contacts or to define trenches inside the diffusion barrier layer or Si wafer underneath the metal-graphene stack. It is noted that graphene obtained can relate to one mono layer, to a bi-layer, to a tri-layer, and to thicker stacks. In view of graphene formation the graphene layer is provided directly on the metal layer, such that both are in contact with one and another .

In a wet etch embodiment then a peroxide (14) is added to the graphene surface during a suitable period, such as of 1-10 min, at a suitable temperature, such as of 10-50 °C, thereby etching the sacrificial metal layer and adhering the graphene layer to the dielectric layer. A liquid as H2O2 is preferred as this is found to improve transport of etch by-products and the etching liquid underneath the graphene. In the method an etch stop layer, such as the dielectric layer, is preferably pre sent under the metal layer. Surprisingly the graphene layer can be adhered directly to the layer underneath the metal layer, by etching the metal layer away. The etch rate of the metal is typically from 0.05-100 nm/sec, preferably 0.1-50 nm/sec, more preferably 0.2-20 nm/sec, such as 0.5-10 nm/sec.

The goal may also be to obtain suspended graphene membranes without requiring any transfer or any wet etching steps in or der to release the graphene membrane. Therein the graphene may be released without wet processing, e.g. in order to create a suspended graphene structure. This can be performed using a gaseous fluorine compound, such as XeF ₂ , for etching. The XeF ₂ reacts with the metal and forms MeF _x , such as MoF ₆ , which is typically volatile. This way the metal will be etched under ^¬ neath the graphene layer, possibly resulting in a suspended graphene layer. The metal acts as a sacrificial layer for dry release using e.g. XeF ₂ .

With the dry etch approach a liquid pull on the membrane is prevented. Therefore the yield of the devices is not reduced due to liquid tensions which may cause collapse of the graphene layer. It is also possible to release graphene membranes with either significantly larger areas or nm-size gaps under neath the membrane. The drawback is that XeF ₂ is a relatively expensive gas.

With the present method a graphene pattern can be defined by patterning the metal layer before graphene CVD, and in ad dition a gap-size between the metal and diffusion barrier layer underneath can be tuned through the metal thickness which may be set during PVD deposition. Virtually no damage to the graphene and no yield-loss, in a simple method are obtained. Graphene is not prone to breaking or collapsing on the surface of the cavity.

The present method provides unique graphene properties, such as a low mass, a high Youngs modulus, a high thermal conductivity, a good optical transparency, and good electrical properties. The present graphene layer may be a very attrac tive material for microelectromechanical systems (MEMS) and nanoelectromechanical systems (NEMS) . Potential applications of suspended graphene membranes are further pressure sensors, gas sensors, ΌND sequencers, resonators, mass measurement devices, micro-fluidics, and lab-on-chip devices. Besides, due to the scalability of the CVD deposition and the completely dry-release method, the

graphene membranes can be scaled to mm or cm-size membranes. Potential applications for these large area membranes are wa ter desalination devices, optical filters, and (hermetic) pro tection membranes of sensitive devices. The graphene layer is adhered to the underlying layer typi cally over a full surface of the underlying layer. Going from small area, typically micrometer scale, graphene layers of the prior art to the present full substrate graphene layers, whilst maintaining or even improving characteristics of the graphene layer, is rather unexpected, also from a physical point of view. For instance overcoming van der Waals-forces over a full cm ² area is very different from pm ² areas. The gra phene layer is crystalline, typically having domains. A size of the domains is in the order of micrometres in cross-section, such as 1-100 pm. On the graphene layer, further layers may be provided, such as to make a product comprising a gra phene layer. The electron mobility of graphene is typically very large, such as larger than 65,000 cm ² /Vs. In a preferred embodiment even 100,000 cm ² /Vs is obtained, e.g. by improving a clean step.

In a second aspect the present invention relates to a prod uct provided by the present method, wherein the graphene layer covers the substrate for 50-100%, such as covering a full sub strate, wherein the at least one graphene layer has less than 1 defect/pm ² , wherein the at least one graphene layer is at least 99.5 % pure, wherein the product is selected from the group of sensors, such as gas sensors, pressure sensors, MEMS, NEMS, IC's, a DNA sequencer, a resonator, an oscillator, a mass measurement device, a micro-fluidic device, a lab-on-chip device, a large area membrane, an optical filter, a hermetic membrane, PV-cells, solar cells, and combinations thereof.

Thereby the present invention provides a solution to one or more of the above mentioned problems and drawbacks.

Advantages of the present description are detailed through out the description.

DETAILED DESCRIPTION OF THE INVENTION

The present invention relates in a first aspect to a method of providing a graphene layer on a substrate.

In an exemplary embodiment the present method may further comprise cleansing the surface with an aqueous solution (15), thereby removing the peroxide, preferably with demi water, preferably with a conductivity of < 1 S/m, preferably < 100 pS/m, more preferably < 50 pS/m, such as < 10 pS/m. The cleansing step supports the adherence of the graphene layer to the under the metal lying layer, typically being the dielec tric layer.

In an exemplary embodiment of the present method the gra ^¬ phene layer covers the wafer for 50-100%, such as covering a full wafer, preferably wherein the graphene layer is transfer free. Surprisingly the graphene layer can be adhered to a large part of the wafer and typically to the full wafer. As such an intact graphene layer can be provided. As there is no need to transfer the graphene layer, from e.g. another location to the wafer, the method may be regarded as providing a transfer free graphene layer.

In an exemplary embodiment of the present method the graphene layer has a thickness of one-fifty atoms, preferably two-forty atoms, more preferably three-twenty atoms, such as four-ten atoms. Each graphene layer may be provided on top of one and another. A few layers of graphene are easily achievable, as well as exactly one layer. When many layers of gra phene are provided etching is slightly hampered and therefore less preferred.

In an exemplary embodiment of the present method the gra phene layer is continuous and/or crystalline. As such the characteristics of the graphene layer can be controlled well and the applicability of graphene therewith increases.

In an exemplary embodiment of the present method at least one layer and/or substrate may be patterned. As such advanced products and devices can be made, e.g. on a pm scale or smaller .

In an exemplary embodiment the present method may further provide a gap under the graphene layer, and may optionally tune the height and width of the gap.

In an exemplary embodiment the present method may further provide at least one clamp for fixing the graphene layer, or part thereof.

In an exemplary embodiment of the present method the diffu sion barrier layer (11) is selected from SiN, Si0 ₂ , and SiC. These dielectric materials can be provided using typical techniques available in the semiconductor industry. It is however preferred to use SiC>2. In an exemplary embodiment of the present method the transition metal layer (12) is a period 4 or 5 metal layer, more preferably a Ni, Co, Cu, Nb, Mo, Fe, or Tc layer, such as Mo, or an alloy thereof. Mo, or alloys is preferred in view of providing good properties for growing graphene and in view of being etched away, while during etching graphene remains intact .

In an exemplary embodiment of the present method the perox ide is selected from hydrogen peroxide, preferably as an aque ous solution, such as a 10-50 vol.% peroxide solution, prefer ably 20-40 vol. %, such as 25-35 vol.%.

In an exemplary embodiment of the present method the die lectric layer has a thickness of 1-1000 nm, preferably 2-500 nm, more preferably 5-250 nm, such as 20-100 n .

In an exemplary embodiment of the present method the metal layer has a thickness of 1-500 n , preferably 5-300 nm, more preferably 10-200 nm, such as 20-100 nm. It is preferred to use a somewhat thicker layer, e.g. of 100 nm, if a lower de composition temperature, e.g. 650 °C, is used. Likewise it is preferred to use a somewhat thinner layer, e.g. of 30 nm, if a higher decomposition temperature, e.g. 1000 °C, is used.

In an exemplary embodiment of the present method the dielectric layer is provided by CVD, ALD, oxidizing, nitride formation, or carbide formation, spinning, or thermally growing.

In an exemplary embodiment of the present method the tran sition metal layer is provided by sputtering, PEVD, CVD, or

ALD.

In an exemplary embodiment of the present method the gra phene layer is grown by CVD.

In an exemplary embodiment of the present method the peroxide is provided as a puddle, such as in a relatively small pool of liquid.

In an exemplary embodiment of the present method demi water is provided to attach the graphene to the dielectric layer.

In an exemplary embodiment of the present method the wafer is dried. Therewith the wafer can be used for further pro cessing, for a final product, etc.

In an exemplary embodiment the present method comprises providing a carbon source, a hydrogen source, and an inert carrier gas for growing graphene, introducing the carbon source into a conditioned environment at a pressure of less than 1000 Torr, such as less than 500 Torr, and at a flow rate of less than 500 seem, such as less than 200 see , such as 1- 10 seem, decomposing the carbon source into at least carbon, synthesizing graphene from carbon upon activation by the metal layer during a predetermined period thereby forming at least one layer of graphene preferably conformally on the metal sur face. Therewith a controlled process for growing graphene is provided .

Typically decomposition of the carbon source takes place at certain process conditions, e.g. temperature, pressure, time, power, etc., whereas synthesis of graphene takes place at other process conditions, e.g. at a lower temperature. As mentioned above synthesis is typically supported by presence of a catalytic material, such as a metal.

It has been found experimentally that graphene can be formed as a conformal layer, i.e. forming a more or less uni form layer with respect to thickness. It is noted that during formation of a crystallographic material, such as graphene, a growth process may involve defects, dislocations and topo graphical effects, such as a slope. In other words, on a microscopic scale some non-uniformity may exist.

As such decomposition and synthesizing preferably take place in a well-conditioned environment, the environment being adaptable in view of required process conditions, and the environment being extremely clean. Even more preferable the en vironment can be used for all or many of the present (optional) method steps. The environment preferably is a vacuum chamber, such as a CVD chamber, a PVD chamber, and combina tions thereof.

It is noted that hydrogen can be added as active gas pre cursor in order to improve deposition of carbon. In general cooling between a step of providing carbon and a step of syn thesizing graphene takes place at a certain rate, such as 10- 50 °C/minute. The plasma can be used to further reduce the synthesis temperature to 500-700 °C. Thereby a well controlla ble graphene layer is provided of excellent quality, e.g. in terms of integrity, dislocation density, electrical proper ties, gas and liquid ( im} permeability, etc. Further the layer can be removed well from the support.

Typically decomposition takes place during a time of 1-600 seconds, such as 5-200 seconds, e.g. 10-50 seconds. Typically synthesis of graphene takes place during a time of 1-500 minutes, such as 2-200 minutes, e.g. 10-30 minutes. Such is considered rather quick and efficient.

In an exemplary embodiment of the present method the carbon source comprises pure C ¹² or comprises pure C ¹³ . Such may be ad vantages in view of characteristics of a graphene obtained by the present method.

In an exemplary embodiment of the present method the carbon source is selected from benzene, naphthalene, toluene, a hydrocarbon preferably having one or more double bounds, such as Ci-i ₈ alkene, such as acetylene, and Ci-is alkane, methane, ethane, ethylene, propane, and combinations thereof. The car bon source may be provided at a partial pressure of 1Q ⁶ -2*10 ⁺⁴ Pa, such as 53-108*10 ² Pa (mbar) at ambient temperature. It is preferred to use benzene, as less energy is needed to decom pose benzene. Having an optional functional group present may improve formation of graphene.

In an exemplary embodiment of the present method the car rier gas is an inert gas, such as a noble gas, such as He, Ne, Ar, Kr, preferably Ar.

In an exemplary embodiment of the present method the hydro gen source is Hz, a hydrogen flow rate is from 0.5-1500 seem, such as 1-1000 seem, e.g. 5-500 seem, or 10-20 seem.

In an exemplary embodiment of the present method prior to providing the graphene layer a surface roughness and/or impu rity level of the metal layer surface is reduced. Therewith the quality of the graphene is improved, larger areas of gra phene can be formed, and the crystallinity is improved. In principle a transition metal may provide the present ad

vantages. However, it may be important to clean a surface of the metal and reduce the surface roughness thereof and/or reduce an impurity level thereof. Surface roughness, often shortened to roughness, is a measure of texture of a surface.

It is typically quantified by vertical deviations of a real surface from an ideal (perfect flat) form. If these deviations are large, the surface is rough; if they are small the surface is smooth. Surface roughness can be determined with a pro- filometer. The present surface is typically flat over a range of about 100 by 100 nm ² . With the present method an especially with Mo a sufficient flatness can be obtained, such as by chemical (pre- ) etching of a surface. The surface may be treated with a diluted carboxylic acid, the carboxylic acid being free of elements other than C, H and 0, such as formic acid, acetic acid, propionic acid, butyric acid, valeric acid, oxalic acid, and benzoic acid. The pH may be from 1.5-6, preferably from 2-5, more preferably from 2.5-3, that is rela tively acidic. Thereby a pristine surface, such as of Mo, is obtained. During treating the surface temperature may be increased, preferably to a temperature of 50-150 °C, more preferably to a temperature of 60-90 °C, such as from 70-80 °C. Thereby a pristine surface, such as of Mo, is obtained. The surface may be treated during a period of time from 5 minutes to 4 hours, preferably from 10 minutes - 1 hour, such as from

15 minutes-30 minutes. Such is found sufficient. A solvent for the acid may be water or an alcohol being free of elements other than C, H and 0, such as methanol and ethanol. The sur face may be annealed at a temperature from (T _meit -200) °C - (Tmei -1) °C, preferably from (T _meit -70) °C - (T _meit -5) °C, more preferably from (T _meit -50) °C - (T _meit -10) °C, such as from

(Tmei _t -20) °C - (Tm _eit -15) °C. Preferably the surface is not heated too much. It has been found that a temperature below the melt temperature and high enough provides the best result e.g. in terms of surface roughness and impurity levels. The surface may be annealed during a period of time from 5 minutes to 4 hours, preferably from 10 minutes - 2 hours, such as from

15 minutes-60 minutes.

In an exemplary embodiment of the present method decompos ing and synthesizing is performed during a period of time from 1 minutes to 24 hours, preferably 2 min-12 hours, more prefer ably 5 min.- 1 hour, such as 10-30 min.

In an exemplary embodiment of the present method decomposing takes place by increasing the temperature in the condi tioned environment, preferably to 400-1600 °C, more preferably to 500-1400 °C, such as to 700-1200 °C.

In an exemplary embodiment of the present method a heating rate is from 10-200 °C/min., preferably 20-100 °C/min., more preferably 30-70 °C/min., such as 40-60 °C/min.

In an exemplary embodiment of the present method a cooling rate is from 10-200 °C/min., preferably 20-100 °C/min., more preferably 30-70 °C/min., such as 40-60 °C/min.

In an exemplary embodiment of the present method the gra phene consists of pure C ¹² or consists of pure C ¹³ , such as by providing pure precursors.

In an exemplary embodiment of the present method the at least one layer has a defect density of less than 10 ^_1 defects/cm ² , preferably less than 10 ^-3 defects/cm ² , more prefer ably less than 5*10 ^~4 defects/cm ² , even more preferably less than 10 ^~4 defect/cm ² . A defect density was also determined and confirmed based on the Raman spectra by comparing peaks thereof and averaging values of fifteen spectra taken over a full wafer.

In an exemplary embodiment of the present method domain sizes are from 50 nm diameter and larger, preferably > 100 nm, more preferably > 500 nm, even more preferably > 1 pm, such as

> 10 pm.

In an exemplary embodiment of the present method the at least one graphene layer is at least 99.5 % pure, preferably at least 99.9 % pure, and in case of C ¹² more preferably at least 99.99 % pure, such as 99.999% pure.

In an exemplary embodiment of the present product the gra phene layer covers the substrate for 50-100%, such as covering a full substrate. Such is orders of magnitude larger than prior art methods provide, and hence prior art product relate to .

In an exemplary embodiment of the present product the product is selected from the group of sensors, such as gas sensors, pressure sensors, MEMS, NEMS, IC' s, a DNA sequencer, a resonator, an oscillator, a mass measurement device, a micro fluidic device, a lab-on-chip device, a large area membrane, an optical filter, a hermetic membrane, PV-cells, solar cells, and combinations thereof.

In an exemplary embodiment the present product may comprise a suspended graphene layer.

In an exemplary embodiment of the present product the gra phene may have a low mass (mass of 1 m ² < 0.8 mg) .

In an exemplary embodiment of the present product the gra phene may have a high Youngs modulus {> 500 GPa, such as 1 TPa) .

In an exemplary embodiment of the present product the gra phene may have a high tensile strength (>100 GPa, such as >

130 GPa.

In an exemplary embodiment of the present product the gra phene may have a high thermal conductivity (1500-2500 W-m ^_1 -K ^_1 ) .

In an exemplary embodiment of the present product the graphene may have good optical transparency (refractive index and extinction coefficient values at 670 nm wavelength are 3.135 and 0.897, respectively, and a transparency of more than 72% at 450 nm, 75.5% at 550nm, gradually increasing to 77% at 700 nm) .

In an exemplary embodiment of the present product the gra phene may have good electrical properties (electron mobility of > 65000 c ² /Vsec, and resistivity <10 ^-6 Qcm, electrical re sistance of <1200 □/□) .

In an exemplary embodiment of the present product the product is an oscillator, and wherein the oscillator frequency may be from 1-20 MHz

In an exemplary embodiment of the present product at least one of the graphene layer may have an area of 1 mm ² -100 cm ² , the graphene layer may be partly suspended, and the gra phene layer may have less than 1 defect/pm ² .

The one or more of the above examples and embodiments may be combined, falling within the scope of the invention.

EXAMPLES

The below relates to examples, which are not limiting in nature .

MATERIALS AND METHODS

Wafer-Scale Transfer Free-Process of Graphene on Mo

For the fabrication process, the starting material is a Si wafer satisfying the following features:

• CZ Si (p-type) wafer purchased from OKMETIC

• Orientation: 1-0-0, 0° of orientation • Resistivity: 2-5 Gem

• Thickness: 525 ± 15 mpi

• Diameter: 100.0 ± 0.2 mm

The Si wafer is then covered by 90 nm of thermally grown Si0 ₂ in a tube furnace provided by Tempress Group.

Afterwards, a layer of Mo, 50 nm thick, is sputtered from a pure (99.95%) Mo target by using a Sigma machine sputtering.

The metal deposition is performed under vacuum by setting the pressure, temperature and target power at 10-7 Torr, 50 °C and 1 kW, respectively.

The graphene growth by Chemical Vapor Deposition takes place in an AIXTRON BlackMagic Pro reactor at 935 °C for 20 min, using Ar/H ₂ /CH ₄ as feedstock at a pressure of 25 bar and a flow rate of CH ₄ equal to 25sccm. The cooling rate after the graphene growth is quoted as 30 °C/min.

In order to etch the Mo underneath graphene, the entire wa fer is thoroughly covered by a puddle of hydrogen peroxide (H2O2) at 31% (vol.) purchased from D-BASF.

The stack formed by Si+ Si0 ₂ +Mo+graphene and H2O2 is then left for 5 minutes.

Finally, de-ionised water having conductivity of about 5 pS/cm is poured out on the H ₂ 0 ₂ puddle aimed to splash H2O2 so that the graphene foil lands on Si0 ₂ .

This invention clearly allows to prove that graphene on large area can be obtained. Also it is also clearly demon strated that the transfer process for CVD grown graphene can be overcome, also achieving much more better results with re spect to what it is usually reported when using a graphene transfer step. It is also proved that graphene material re mains unchanged, having the same properties on the Si0 ₂ layer compared to what it had as grown on the Mo layer. The present method is extremely simple, fast, reliable and environmentally friendly and represents a crucial improvement with respect to the state of the art.

It is considered totally not obvious that an entire layer of graphene would actually land on the entire wafer after the Mo layer removal. After the Mo layer etching underneath graphene, by H2O2, the puddle of H2O2 is removed by de-ionised water and the entire film of graphene completely lays down on the Si wafer covered with SiCk. On the opposite, what could be expected is that the foil of graphene could be etched away while etching Mo underneath graphene itself. Strictly joined to this phenomenon, the other no-obvious feature is that gra phene remains completely undamaged and continuous while land ing on the wafer without Mo. In fact, due to the nanometric thickness of the foil and the area of hundreds cm ² , it could be expected that the film

could be broken on the wafer. The surface roughness of the Mo was determined with AFM before and after etching and is found to reduce from 5.3 nm to 3.1 n on a 25 pm ² area.

In an example a relatively low temperature of 300 °C -1100 °C, such as 650 °C, is used to decompose the carbon source. In a further example a 100 mm or 200 m (diameter) (Si-) wafer support is used, having a thickness of 150-1500 pm, preferably 250-750 pm. It has been found that graphene grows uniformly and covers structures well through conformal growth. It has been found that also larger layers can be made, such as with a 300 mm Si-wafer, the layers typically being continuous. In principle size of the graphene layer of the invention is lim ited by the size of a chamber. Typical dimensions of the pre sent graphene layers are 1-300 mm long, e.g. 2-150 , such as 5-50 mm, 1-300 mm wide, e.g. 2-150 mm, such as 5-50 m, and 1- 10 layers thick, e.g. 2-5 layers. The number of layers depends amongst others on a time of synthesizing and a thickness of the sorbent. More sorbent (volume) provides an opportunity of growing thicker graphene. Typically the layer obtained is of good quality and homogeneous. Such relatively large layers can be applied as such in e.g. a device, or in parts thereof, e.g. a repetition of structures may be grown.

By introducing the carbon source such as benzene or methane gas into a sputter chamber as feed stock there is no need for breaking vacuum. Further there is no need to provide a sol vent. Thus a physical vapour deposition system can be combined with a chemical vapour deposition to achieve unique results.

By directly synthesizing graphene after a metal, such as Mo, sputter deposition without breaking the vacuum the quality of graphene is improved significantly, the contamination is re duced, and the efficiency of synthesis is improved. A further advantage is that the present sputter deposition produces a uniform Mo film is obtained.

A heating stage was used to raise the temperature to about 650 °C. A benzene gas was introduced into chamber during about 5 hours. In a further example the benzene was introduced during 10 minutes. Typically time, flow and temperature are adjusted such that sufficient carbon is being formed for a required layer. Such can be determined by standard tests.

Addition of e.g. benzene may be controlled with high accu racy by using a needle valve to adjust the gas flow rate precisely. A needle valve typically has a relatively small orifice with a long, tapered seat, and a needle-shaped plunger, on the end of a screw, which exactly fits this seat. As the screw may be turned and the plunger retracted, flow between the seat and the plunger is possible; however, until the plunger is completely retracted the fluid flow is significantly impeded. Since it takes many turns of the fine-threaded screw to retract the plunger, precise regulation of the flow rate is possible.

The invention is further detailed by the accompanying fig ures, which are exemplary and explanatory of nature and are not limiting the scope of the invention. To the person skilled in the art it may be clear that many variants, being obvious or not, may be conceivable falling within the scope of protec tion, defined by the present claims.

FIGURES

The invention although described in detailed explanatory context may be best understood in conjunction with the accompanying figures.

Figs, la-f show schematics of the present method.

Figures 2a-b show results of Raman measurements.

Raman spectra.

Figs. 3a-b show AFM recordings.

Figs. 4a-b show schematics of a perovskite solar cell.

DETAILED DESCRIPTION OF THE FIGURES

Figure la schematically shows a silicon wafer 10. On the silicon wafer 10 a dielectric layer 11, such as a Si0 ₂ layer, is provided. On the dielectric layer 11 a transition metal layer 12, such as Mo, is provided. On the metal layer 12 a layer of graphene 13 is grown. In figure 2b on the graphene layer a peroxide 14 is provided. The peroxide layer etches away the metal layer 12, as is shown in figure lc. Then on the peroxide layer water 15 is provided, thereby further adhering the graphene layer to the dielectric layer (fig. Id) . In figure le the water and peroxide are removed, leaving a silicon wafer, a dielectric layer and a graphene layer. In figure If a fluorine etch is used to create a suspending layer of graphene, i.e. a graphene layer bridging an underlying gap 18 and hanging over said gap from one end thereof to another end thereof, typically opposing ends of the gap. The height and width of the gap may be tuned. The graphene layer may be fixed by one or more clamps 17, for instance a ring-shaped clamp, or two opposing L-shaped clamps, which clamps may be integrated in the device. The gap does not necessarily have to go through the silicon wafer 10, but can also only be created in the die lectric 11 or even by only removing the metal catalyst 12.

Figures 2a-b show results of Raman measurements. In figure 2a a final wafer is shown with a graphene layer on top. The dots illustrate where Raman measurements were performed. Fig ure 2b shows the results of the Raman measurements, with on the vertical axis the intensity, and on the horizontal axis the Raman shift in cm ^-1 . For all locations on the wafer a fully consistent result is obtained, indicating that graphene is formed. The peaks are indicative for the sp2 carbon-carbon bonds. It can also be observed that the graphene is relatively pure .

In fig. 3a an AFM image is formed. As can be seen domains of graphene are formed, having sizes in the order of micrometres. The height of the graphene varies somewhat, partly due to a number of graphene layers being formed locally, as well as due to surface roughness of the underlying dielectric layer and possible incomplete etching leaving some metal residues locally. Fig. 3b shows a similar image at a different location .

For PV-cells two stacks of layers are considered specifi cally, relating to a so-called inverted and standard perov- skite devices (see fig. 4a-b) . The substrate 10 therein is typically glass or silicon. In the in-verted device the anode 21 is an optically transparent electrode, such as FTO or the present graphene, in contact with the substrate. In contact with the anode and the perovskite 31 is the present hole transport layer 41. A metal layer 22 (such as Al or graphene) acts as second electrode. Typically an intermediate layer 51, such as a fullerene, is present between the cathode and perov skite. The perovskite may be CH3NH3Pbå3, CH3NH ₃ PbBr ₃ , a mixed halide, such as CH3 H ₃ PbX]X2, (Xi and X ₂ being independently se lected from halides, wherein three halides are present), such as CH ₃ NH ₃ Pbl3-xCl _x , and an inorganic perovskite, such as CsSnX ₃ , or mixed inorganic perovskite, such as CsSnXi3- _x X2 _x (X, Xi and X _å being independently selected from halides), such as iodides.

In the standard perovskite device the cathode 21, such as FTO is on the substrate 10, such as glass. A metal layer (such as Au) acts as second electrode 22. In contact with the cathode and perovskite layer 31 is an intermediate layer 51, such as TiC>2. In contact with the anode 22 and the perovskite 31 is the present hole transport layer 41. In addition a passivation layer may be provided, and further intermediate layers may be present.

The figures have been detailed throughout the description.