FI127861B - A method and a device for dynamic generation of hot electrons into aqueous solutions - Google Patents

Abstract

Methods and devices are disclosed to produce hot electrons in controlled ways and to introduce homogenous flux of hot electrons into aqueous or other electrically conductive liquid phases to carry out specific chemical reactions which are only possible with the hot electrons.

Description

A METHOD AND A DEVICE FOR DYNAMIC GENERATION OF HOT ELECTRONS INTO AQUEOUS SOLUTIONS

FIELD OF INVENTION

The present invention deals with electronic engineering and electrochemical reactions by highly energetic or hot electrons which reactions are generated in specific conditions on electrical components. In particular, the present invention aims at producing hot electrons in controlled ways and introduces them into aqueous or other electrically conductive liquid phases to carry out specific chemical reactions which are only possible with the hot electrons.

BACKGROUND OF THE INVENTION

Hot electrons are exceptionally high-energy electrons occurring in certain rare situations on electrical components. Hot electrons are characterized by high local effective temperature and they can travel across various energy barriers, in particular, tunnel through electrically resistive layers. This phenomenon is widely known in applications in microelectronics like MOSFET (metal-oxide-semiconductor field-effect transistor), field effect transistors (FET), non-volatile memory devices (US 9,368,209 B2 /2016 Miiani et al.), electrically-programmable read-only-memory (EPROM) and a flash memory cells (US6327187B1 Bergemont et al.), and superconducting transistors (US 5,318,952 Tsunehiro Hato). Basically, hot electrons are used for charge transfer between structures of a device separated by a barrier. Dynamics of hot electrons is described by Maxwellian electrodynamics and quantum physics. These theories allow knowing averaged characteristics of the free hot electron gas and therefore require higher local temperature. Often, like in the nonvolatile memory devices, only the charge transfer is required. In such applications therefore, it is not essential for the result what is happening during travelling of hot electrons. However, to drive specific chemical reactions the hot electrons must be in rather narrow energy limits. In some applications concerning especially aqueous solutions, it is important that the pathways and energies are strictly controlled, exemplified by the participation of hot electrons in excitation processes in electrochemiluminescence (ECL) from luminophores in solutions (US 6,479,233 B1 Bard et al., US 6,517,777 B2 Liljestrand et al., US 6,808,939 B2 Sigal et al., US 8,211,279 B2 Bard et al., US 8,920,718 B2 Kulmala et al., US 9,075,018 B2 Geddes

20175624 prh 01 -04- 2019 et al., US 9,139,766 B2 9/2015 Qian et al., and others). Dynamic characteristics of hot electrons have thereby a strong impact onto the chemical and photonic processes. A wide-energy spectrum of hot electrons will result in large noise and low signal-to-noise ratio because the excess of energy will transfer to heat and to lower energy states than required and result in non-radiative transitions and consequently low efficiency of light emission. On the other hand, ECL is an excellent means to monitor hot electrons and the processes leading to non-radiative and radiative transitions. The generation of hot electrons can be proved by applying fluorescent compounds which in photo-excitation are excited in the UV region of electronic spectrum.

In controlled release of hot electrons, there are several drawbacks in the prior art. While hot electrons have been studied quite largely from the viewpoint of their existence in solid state physics, these studies have focused primarily on avoiding their harmful effects onto solid materials by destroying semiconductor structures. There is a need to improve the control of producing homogenous flux of hot electrons into solution phase from solid structures. This present disclosure specifically addresses this question and the approach is completely different than previously set forth in solid state physics. The invention described in this disclosure provides significant improvements to the known art, and solves flaws of the currently used methods and devices.

SUMMARY OF THE INVENTION

The present disclosure describes significantly improved traits providing devices and methods to inject intensive and constant flow of hot electrons into liquid phase by exploiting sensing and feedback systems.

It is an object of this invention to improve the chemical reactions by hot electrons with better control of chemical composition and improved control of hot-electrons distribution in three dimensions.

It is an object of this invention to provide a three dimensional control of the flow of hot 30 electrons and thereby provide an optimal excitation method for each case with a computer program under a feedback from the electrically active layer and/or luminescence signals.

20175624 prh 01 -04- 2019

It is an object of this invention to provide a method and device to dynamically optimize average energy of hot electrons.

It is an object of this invention to provide significant improvements in performance of ECL devices by the precise dynamic control of injection of hot electrons into the electrically active layer containing the chemically reacting molecules.

It is an object of this invention to provide decreased size of the devices, increased number of materials which can be used in the production of the devices, increased number of label compounds, increased number of functional parameters, and significantly improved sensitivity of detection through dynamically controlled and optimized generation of hot electrons.

While the present disclosure can be applied in various fields of chemistry to carry out chemical reactions, one of the current potential commercial applications of hot electrons is their use in diagnostics, particularly, in electro-chemiluminescent (ECL) detection of biomolecules. Accordingly, it is an object of this invention to provide improved methods and devices for use in diagnostics and electro chemiluminescent detection of biomolecules by means of dynamically controlled and optimized generation of hot electrons.

The ECL detection normally exploits label substances which can be excited electrochemically to produce luminescence which is correlated to quantitative amount of an analyte in aqueous solution. As very small amounts of chemically reactive labels must be detected quantitatively the ECL system has exceptionally high demands for stability and reproducibility. The present disclosure describes novel ways of how constant flux of hot electrons of determined energy values can be created and injected into solutions.

It is an object of this invention to provide a device for generating a homogenous flux of hot electrons in an aqueous solution, said device comprising a cathode electrode and at least a first and a second control electrodes on a substrate surface; a barrier layer separating an anode electrode from the substrate, the barrier layer being made on the cathode electrode, an isolation layer separating the control electrodes from the anode electrode; and an aqueous sensing layer above the barrier layer; all of said electrodes being connected to an electronic circuitry; the control electrodes

20175624 prh 01 -04- 2019 creating a lateral electro-static field and the anode providing a vertical electromagnetic field across the barrier layer; wherein hot electrons are generated by electric cathodic pulses, and the hot electrons gain kinetic energy by moving laterally in the lateral electrostatic-field between the first and the second control electrode, and wherein the lateral movement is changed to a vertical movement by cathodic voltage pulses providing change of lateral momentum of an electron to vertical momentum, whereby the hot electrons with sufficient energy move through the barrier layer via quantum tunneling and an energetically homogenous flux of hot electrons is generated into the aqueous sensing layer.

It is an object of this invention to provide a method to generate a homogenous flux of hot electrons in an aqueous solution, said method comprising: a) providing a cathode electrode and at least a first and a second control electrodes on a substrate surface; b) providing an anode electrode; c) providing a barrier layer on the cathode electrode to separate the anode electrode from the substrate, and an isolation layer to separate the control electrodes from the anode electrode, and an aqueous sensing layer above the barrier layer; d) connecting all of said electrodes to an electronic circuitry; e) creating a lateral electro-static field between the control electrodes, and a vertical electro-magnetic field across the barrier layer by the anode electrode; f) applying electric cathodic pulses to generate hot electrons at the cathode, g) allowing the hot electrons to gain kinetic energy by moving laterally in the lateral electrostatic-field between the first and the second control electrode, h) changing the lateral movement to a vertical movement by cathodic voltage pulses, thereby causing the hot electrons with sufficient energy to move through the barrier layer via quantum tunneling and generating a homogenous flux of hot electrons into the aqueous sensing layer. The method includes a feedback correlating photonic process and electronic process and the feedback times are shorter than ECL decay times.

SHORT DESCRIPTION OF THE DRAWINGS

FIG. 1. A prior art MOSFET (metal-oxide-semiconductor field-effect transistor) transistor. A schematic drawing of a MOSFET transistor comprising a source, a drain, a gain and back electrodes. The device is made by using a CMOS (Complementary metal-oxide-semiconductor) fabrication processes creating n+

20175624 prh 01 -04- 2019 regions for source and drain contacts in a substrate, usually p-type silicon, gate oxide layer separating the substrate from the gate contact which is usually made of polysilicon. There are also depletion regions surrounding n+ regions in p-type substrate.

FIG. 2.a, b and c. Fig 2 a) illustrates a schematic drawing of a device according to this disclosure for generation of hot electrons. The device comprises a substrate 51, electrodes 53, 52 and 54 prepared near the surface of the substrate, a barrier layer 55, top electrode 57 separated from the barrier layer by an aqueous sensing layer 56 which is filled through channel 58. The electrodes 52, 53, and 54 are separated from the top electrode 57 by an isolation layer 61. Fig. 2b) and Fig. 2c) show two regimes of generation of hot electrons. In accordance to Fig. 2b, hot electrons are generated in the electrically active sensing layer 56 by electrical pulses generated along with the vertical axis, shown by arrow 1 - 2, across the barrier layer 55. Fig. 2c shows the generation process of hot electrons when the hot carriers first gain their energies during lateral drifting shown by arrow 1 - 2 followed by additional impact of a pulsed voltage in vertical direction arrow 2-3 resulting in changing the momentum of the carriers adding energy to the hot electrons to enable tunneling through the barrier layer.

FIG. 3 illustrates the cross sectional view of the device in Fig. 2. The device can be fabricated with semiconductor processes like CMOS (Complementary metal-oxidesemiconductor) and SoC (system on a chip) methods. Substrate 51 can be made of, for example, of p-type silicon. The electrode contacts 52, 53, and 54 can be made with optical lithography and plasma etching followed by deposition of contact material, for example, n-type silicon or a metal. The metal can be deposited by evaporation in vacuum or magnetron sputtering followed by polishing. A resistive or barrier layer 55 is the layer which defines the energy barrier allowing transfer of only those hot electrons having sufficient high energies into the sensing layer 56. The resistive barrier layer 55 is exemplified by a silicon oxide film having thickness of few nanometers. A top electrode 57 is separated from the resistive barrier layer 55 by an air gap 69 which is capable of being filled with the aqueous solution to form a sensing layer and is connected to the substrate with an anchor 66. There can be more than one anchor. The top electrode 57 can be made of a polysilicon or transparent conductor. The sensing layer 56, for example an electrolyte, is placed in

20175624 prh 01 -04- 2019 between the resistive layer 55 and the top electrode 57. The conductive regions are electrically isolated by an isolation layer 61. Also in this figure the top electrode has a hole 80 in the center for observation of photon radiation produced during ECL. The device can have one or more such holes.

FIG. 4 illustrates an embodiment where a device according to this disclosure involves several thin layer structures that provide multiple electrical connections 101, 102, 103, 105, 106, 107, barrier layer108, an air-gap between the electrodes 106, 107, 108 at the bottom and the electrodes 102, 103 and 105 on the top which is filled with aqueous solution through the channel or holes 116 to form a sensing layer 115 Also liquid samples may be added into the sensing layer through the channel or holes 116 . Electrode contact 107 provides AC or DC+AC voltage through the barrier layer 108 and sensing layer 115. Electrode contact 106 is for cathodic pulse excitation of hot electrons into the sensing layer. Sensing layer 115 comprises active electrochemical materials, such as luminophores and thus provides ECL after electrical excitation.

Sensing layer 115 also provides a reaction milieu for ligands to be excited and place to read the tunneling current through the barrier layer. Element numbers 110, 111 are not mechanical structures but indications of electron charge density near surfaces which impact the tunneling hot electrons. Higher electric field allows higher probabilities for electron tunneling process. The top electrode 102 can be deformed due to electrostatic forces. The illustration of the figure is not, however, in real scale and the bending as illustrated in the figure is not visible in reality. In spite of that, even very tiny deformation can strongly increase probability of electron tunneling. With no electrostatic forces (no voltages), the top electrode 102 is straight in the horizontal plane.

FIG. 5a, b, c, d show different regimes of the tunneling hot electrons in the vicinity of the cathode124 and anode 121 contacts. Element numbers 122, 123 are not mechanical structures but are related with density of electronic states and electrostatic field. The closer the features 122, 123 are to each other, the bigger probability there is for the tunneling. Fig. 5a shows no tunneling; the gap between

122 and 123 regions is very large. Fig. 5b shows the tunneling through one channel (flow of electrons FE1), the regions 122 and 123 are closer to each other due to increase electrostatic field (increased voltage). Fig. 5c reveals few tunneling channels (flow of electrons FE2), regions 122 and 123 are even more closer due to

20175624 prh 01 -04- 2019 further increase of the electro-static field and Fig. 5d exhibits multichannel tunneling 134c where regions 122 and 123 are overlapped. Different numbers of hot electrons are injected which number varies from 0 (Fig. 5a) to few (Fig. 5b), a few (Fig. 5c) and many (Fig. 5d).

FIG. 6 illustrates a schematic diagram of the present disclosure as used for creating an measuring electrochemiluminescence. The device comprises a multilayer structure (shown in middle of the diagram) where the layers are electrically connected to each other and the device is providing controlled flux of hot electrons into the sensing layer. The hot electrons interact with ligands in the sensing layer resulting in radiation of photons. The photon spectrum in the sensing layer is measured by a photo-detector and time-resolved spectroscopy data is obtained. The tunneling current, photon spectrum, and the integrated signal are used for feedback mechanism. The feedback controls are voltage amplitudes, pulse duration, switching time, integration time, and frequency, and they are adjustable parameters. The system incorporates both, electrical and photonic measurements. Both phenomena depend on one another: photon emission depends on the number of hot electrons generated in the very thin sensing layer, about 200 nm, above the oxide film (barrier layer). The tunneling currents indicate electro-dynamics of the process. AC voltage is used to generate fluctuations of charges and carriers that can result in resonant electrochemical reactions at which the ECL process occurs fast. This can happen at a certain resonant frequency which depends on materials and geometry. The effect can be described using effective cross section of molecules and ligands. At resonance, the effective cross section is increased resulting in faster reactions with hot electrons.

FIG. 7. An example of an ECL device is shown. The device comprises substrate 72 of the chip, cathode electrode 73 for electrical pulse excitation, multiple/plurality of electrodes 71 around the cathode separated by the tunneling region (shown in Fig. 5), central electrode 74 for exciting variable electromagnetic field. In a simplest case, AC voltage is applied to the central electrode 74. The shape of the voltage can have a complex form including amplitude modulation, sweeping frequency, or forms described by mathematical expressions. Central region (area around the electrode 74) is reserved for aqueous solution which is usually chemical composition with precisely defined ligands. Because one can have several reading electrodes 76, it is

20175624 prh 01 -04- 2019 possible control tunneling process locally. The tunneling current can be integrated one which allows comparing of the tunneling currents at different locations. Interconnections 75 are for electrical connections of the electrodes and contact pads 75 provide connection to electronic circuitry.

FIG. 8 illustrates another embodiment of a device according to this disclosure consisting of several sensors 9, buffer micro-cavities 12, micro-pumps 15, aqueous tanks 21 connected with tubes 17 and 25 to the sensors 9. The aqueous connection system can be supplied with valves 31 and 32. Electrodes 5 measure tunneling currents, while other electrodes 10, 11 provide electrical connections described in

Figs 6 and 7.

FIG. 9a, b illustrates an ECL device integrated to an optical system. Fig. 9A is a top view of the device and Fig. 9b is a cross section in vertical plane. The device comprises several sensors 149 integrated on a micro-chip 161 (shown in 9b only), micro-pumps 156, tanks 151 for liquids, supplied with micro-valves 145, 146, buffer tanks 155 for liquids and other components supplying power and electronic circuit

159 (shown in 9b only). The chip is placed in a holder 157 (shown in 9b only) fixed in the cylinder 150 (shown in 9b only). The cylinder is mounted on a base 158 (shown in 9b only) which is provided by electrical connectors, bus for electronics and data transfer and other cables. The device is mounted in cylinder 150 covered by flange

142 which can be rotated around vertical axis. There is an optical detection device

160 (shown in 9b only), which is photomultiplier tube, CCD or other related unit, mounted on the flange.

FIG. 10. The band-gap functions are demonstrated for a) 1 eV, b) 2 eV and c) 3 eV with the device of Fig 3. The curves are obtained from tunneling current-voltage characteristics.

FIG.11 a, b. Time-resolved photonic spectra of the chemiluminesence device of Fig. 3 are revealed. The curves of Fig. 11A demonstrate few examples of the spectra at a) 10 ps, b) 1 ms, c) 3 ms, and d) 5 ms after the excitation. Fig. 11B shows decay of light intensity vs. time. The light intensity is shown in arbitrary units (a.u.). Time is shown in milliseconds (ms).

FIG. 12 a, b. Time-resolved photonic spectra of the chemiluminesence device are shown. Fig. 12A shows spectra in a different logarithmic scale. Fig. 12B shows an example of the spectrum in the linear scale. The times for a), b), c) and d) are the same as in Fig. 11.

FIG. 13. This figure illustrates time-resolved photonic spectra of the ECL device of

Fig.3 for a different chemical aqueous solution. Fig. 13A shows few examples of the spectra at a) 10 ps, b) 1 ms, c) 3 ms, d) 5 ms and e) 10 ms after the excitation. Fig. 13B shows time-dependence of light intensity (in arbitrary units, a.u.). The decay time I lifetime is different than that of the Fig. 11.

DETAILED DESCRIPTION OF THE INVENTION

20175624 prh 01 -04- 2019

Definitions

1. Sensing layer. A thin aqueous layer above the oxide film where hot-electrons are generated, allowing diffusion of hot electrons inside the layer during relatively long time so that these hot electrons would most probably take part in electro-chemical reactions. The sensing layer comprises active electrochemical materials, for example, bio assays and ligands in aqueous solutions producing photon emission during electro-chemical reactions. Depending on the context, also terms electrically active layer or aqueous layer are used interchangeably with the sensing layer.

2. Barrier layer A resistive or barrier layer is the layer which defines the energy barrier allowing transfer of only those hot electrons having sufficient high energies into the sensing layer. The barrier layer is exemplified by a silicon oxide film having thickness of few nanometers.

3. Valence band. The highest energy band occupied by electrons in a solid state crystal.

4. Conduction band. The energy band above the valence band. It is only partially filled by electrons and higher energies for electrons are required to occupy this

20175624 prh 01 -04- 2019 band. The conduction band is separated from the valence band by an energy band gap

5. Energy Band gap. The minimum energy difference between the top of the valence band and the bottom of the conduction band.

6. Energy band gap structure. In the devices disclosed here, the barrier layer is an energy band gap structure.

7. Fermi level. An energy level indicating the upper limit of energies of electrons at absolute zero temperature.

Few different mechanisms of hot electron injection associated with the MOS (metal oxide seminconductor) devices are known from solid state physics. These mechanisms include channel hot electron injection, substrate hot electron injection, drain avalanche hot carrier injection, and secondary-generated hot electron injection.

In devices applying these mechanisms, the electrons travel from source to drain along an oxide layer which is biased by a gate voltage. In the case of channel hot electron injection, the gate and drain voltages exceed the source voltage and due to high gate voltage, channel carriers are driven to the gate oxide before reaching the drain. In the substrate hot electron injection, the substrate bias is high, and the 20 carriers gain high kinetic energy in the interface with the oxide layer, and further to the depletion region. Due to high kinetic energy, carriers can be injected into the gate oxide. In the case of drain avalanche hot carrier injection, the drain voltage is higher than the gate voltage, and the carriers are accelerated towards drain depletion layer. The carriers can be injected into the gate oxide. This mechanism is also called 25 impact ionization with formation of electron-hole pairs. Trapping of hot electrons creates a space charge which impacts to the threshold voltage and mobility characteristics. In the case of secondary-generated hot electron injection, the hot carriers are generated from impact ionization with participation of a secondary carrier which was generated before this stage. This happens at high drain voltage which is 30 larger than the gate voltage. Hot electrons, electrons and holes, are generated in

MOS and MOSFET transistors (such as shown in Figure 1 for example) as a harmful side effect. This means that hot electrons are not part of function of these devices. Hence, hot electrons in these devices results in damage of the devices. Therefore,

20175624 prh 01 -04- 2019 the goal has been to avoid generation of hot electrons by developing different designs.

Dynamics of hot electrons drifting along the oxide layer has been investigated in detail. In the method and devices presented here hot electron drifting along the oxide 5 layer is not of interest but a completely different approach is employed here. In this disclosure, generation of hot electrons, the hot carriers, are generated not by drifting along the oxide layer, but by drifting across the oxide layer into the aqueous layer. On the contrary to the known art applying solid state physics, the present invention exploits hot electrons as an operational part of the devices and provides surprising 10 results: The electric carriers gain kinetic energy similarly as any hot electrons, but transit their energy into the electrically active liquid layer through the barrier layer.

According to the present disclosure, there are two formal pathways to generate hot electrons into the electrically active layer: a) a short pathway and b) a two-step 15 pathway. In the short pathway (a), the hot electrons move across the oxide surface layer perpendicularly (see Fig. 2 a). Advantage of short pathway is that travelling distances are short and consequently tunneling time is short. Disadvantage of the short pathway is that a high electrostatic field is required to accelerate hot electrons within this short distance. In the two-step mode (b), hot electrons first drift along the 20 oxide barrier layer and gain high kinetic energy before their direction of movement is changed by a deflection voltage (see Fig. 2b). In this case the electrons need two electrostatic components: the first one is lateral while the second one is vertical. The vertical field is switched on when the hot electrons are accelerated to a required speed. Advantage of this regime is that hot electrons can reach substantially higher 25 energies and can be regulated more exactly. These features are especially useful in developing electrochemiluminescence devises for detection of biomolecules of an analyte in aqueous solution.

Principles related to the invention

Below are described some principles that affect how the devices according to this invention can be developed. A skilled artisan would understand that there are many

20175624 prh 01 -04- 2019 other principles affecting the development and designs and would be able to apply such principles without specific description of such principle here.

Electrodynamics of electrons can be divided as thermionic emission, diffusion, drift, and tunneling. Drift velocities are obtained when the charged carriers are accelerated by an electric field. With no electric field, the carriers move randomly due to the thermal excitation. This results in scattering of the carriers. When an electric field is applied, the carriers move ballistic between randomly distributed scattering events resulting in a flow of electric current. In the present disclosure, the momentum of each hot electron is changed due to additional pulsed voltage which is switched on periodically. The period between the pulses is shorter than the average time needed for carriers to reach the opposite electrode on the same plain so that the hot carriers tunnel through the oxide layer. Thus, lateral electric current is minimal, meaning that most of the hot electrons go through the oxide layer. The scattering depends on defects in the crystal, phonons, thermal phenomena etc. Electrical currents of electrons and holes are opposite in the same electric field. Since electrons and holes have different effective masses, they are characterized by different drift velocities.

Diffusion is an associated phenomenon in the present disclosure because the concentration of the charge carriers varies. The diffusion is due to a gradient in the carrier concentration and is driven by thermal movement. The diffusion results in lowering the gradient of the concentration.

Generation of hot electrons can be described as follows: 1) Tunneling of hot25 electrons is a quantum mechanical effect of electron tunneling through a relatively thin barrier. The number of hot electrons that tunnel through a barrier depends on the thickness of the oxide layer. The tunneling current decreases exponentially when the layer thickness grows. 2) Thermionic emission is heat-induced flow of hot electrons over a barrier when the thermal energy of the hot electrons is larger than the barrier energy. When a metal layer and a semiconductor layer are in contact, electrons flow from the material with lower work-function or higher Fermi energy to the material with the higher work-function or lower Fermi energy. The carrier moved due to thermionic emission, causes bending in the semiconductor band. Each of these two effects plays an essential role as the hot electrons gain a large kinetic energy. Hence, they

20175624 prh 01 -04- 2019 are incorporated phenomena: the bigger the drift, the higher local temperature and thermionicemission and higher tunneling probability.

In the most applications where hot electrons react with molecules in a sensing layer, the number of the molecules in liquid is usually variable. When hot electrons are applied to drive a chemical reaction in a batch process or fixed volume, the amount of starting materials and end products are constantly changing. To maintain optimal reaction, the number of hot electrons and their energies shall be adjusted accordingly.

In order to generate hot electrons in a specific energy region, a resistive layer should have a precise band gap value. For example, in the case of S1O2 barrier, the film must be extremely thin. If the film is too thick (more than 50 nm), the band gap value is too high, and no tunneling will occur. If the film is too thin (below 1 nm), harmful leakage currents occur and hot electrons are not produced. Hence, an ultrathin flawless layer of few nanometers thickness is often optimal. However, fabrication processes of the resistor layers or barrier produce thicknesses distributed around some average value and the band gaps will be distributed around some average value, too. In order to provide an effective tunneling process, methods of scanning tunneling spectroscopy were found useful in the present disclosure to measure an effective band gap value. The disclosed devices have two or more tunneling current components: the tunneling current component 1) through the resistive layer, and the tunneling current component 2) through the resistive layer and the liquid phase in sensing layer. According to the present disclosure these components can be measured and compared for adjustment of the applied voltages.

The total band gap consists of the band gap of the resistive barrier layer and the effective band gap of the sensing layer. Thus, if the sensing layer is empty, no hot electrons will be generated. Therefore, properties of the sensing layer affect the generation of hot electrons as well as the thickness of the resistive barrier layer. In accordance to the present disclosure, the effective band gap of the sensing layer is less than the band gap of the resistive barrier layer. In accordance to another embodiment, the effective band gap of the sensing layer is close to the value of the band gap of the resistive layer. The upper contact above the aqueous layer can be

20175624 prh 01 -04- 2019 deformed due to electrostatic force and the thickness of the sensing layer is smaller beneath the deformation area (c.f. Figure 4). Accordingly, the effective band gap of the aqueous layer is smaller here. Thus, the total effective band gap of the device can be adjusted to obtain the required number of hot electrons. Figure 5 a-d shows different regimes of the tunneling hot electrons in the vicinity of deformation of the upper electrode. The first figure (5a) shows no tunneling, the second figure (5b) shows the tunneling through one channel, the third figure (5c) shows few tunneling channels and the last figure (5d) shows multichannel tunneling. Different numbers of hot electrons are injected started from zero, few and many.

The aqueous solution can comprise charged particles, carriers including holes, which can interact with the injected hot electrons. This property can be used, for example, for detection impurities in the aqueous sensing layer, generation of light, recharging liquids and others. Even there are many applications of hot electrons to carry out unique chemical reactions in conductive liquids, aqueous liquids, or other electrically conductive phases, the present disclosure is mainly illustrated here by the ECL because of its easy to detect even visually in a dark room, or with a camera. However, many other hot electron -driven chemical reactions are known from the scientific literature, and the methods and devices disclosed here may be useful in such reactions as well.

The hot electron -induced luminescence can be especially well exploited in diagnostic equipment and sensors used for medical and biological applications. Usually such systems are constructed to have label compounds functioning as indicators of substances to be analyzed. The label compounds are excited by hot electrons to produce luminescence which is detected and analyzed. ECL is the result of multistep chemical reactions with energy release in the form of photons. Previous studies on ECL focused primarily on materials in ECL reactions and not on the optimal design of producing controlled hot electron flows into the electrically active sensing layer or on the location of the biomolecules to be analyzed.

ECL can be divided into two principles, those using normal fluorescent labels and those using long-decay-time luminescence labels which can exploit the so-called time-resolved detection technologies. On the other hand, ECL can be divided into the anodic ECL using normal electrical current at anode and the cathodic ECL using hot

20175624 prh 01 -04- 2019 electrons only known to occur at cathode (termed as CECL). Time -resolved technology is normally only possible with CECL with the luminescent labels having the long light decay component. Time-resolved technology using pulsed excitation enables the increase of the detection sensitivity 10-50 -fold compared to the fluorescent technologies as shown by photoexcitation and therefore the present disclosure focuses, but is not limited, to this time-resolved technology. CECL can apply a number of fluorescent and luminescent compounds known from photoexcitation methods like Fluorenylmethyloxycarbonyl (FMOC) -, or anilinenaphtalene sulphonic acid (ANS)-based methods, or chelates of lanthanide ions including Tb(lll) ion (emission spectrum around 550 nm), Ru(bpy)3²⁺ (emission around 610 nm), Yb(lll)-L1 (emission around 980 nm) and others. CECL is based on creation of excited states of such luminescent label compounds by high energy hot electrons in electrically active layer through electrical pulsed and multistep-electron transfers during each excitation cycle. Eventually, the excited chelates can emit measurable light. Physics of ECL have many similarities with photoluminescence but not all labels are interchangeable between these principles. It is noteworthy that only hot electrons can trigger luminescent labels in aqueous phase. Hot electrons can fly ballistic about 100 times longer from the surface compared to the normal electrons. Therefore the physico-chemical conditions on the electrically active layer are not comparable in the anodic and cathodic techniques.

Despite many advantages of ECL, there are a few severe physical drawbacks: i) in order to get high sensitivity, the sensing device must have relatively large surface area, ii) luminescing decay curves have relatively big noise component, and iii) the fabrication processes of the devices producing hot electrons into solutions is challenging since several parameters affect the quality of the resistor surfaces. Therefore the density of the flow of hot electrons can vary on the surface from place to place. Furthermore, the variation is three dimensional also in vertical direction due to the different energies of hot electrons.

The physical drawbacks must be fitted against the chemical drawbacks in the electrically active layer. In the diagnostics systems which exploit, for example, bioaffinity, the chemical milieu is highly complex. The size of coating antibodies, for example, are of dimension of about 15 nm, analyte may be of the same size, and the detection antibody with the label compound is also of that size. Therefore, the

20175624 prh 01 -04- 2019 measured label compounds locate roughly in the area 30-45 nm from the surface. The maximum trajectory of hot electrons is up to 100-200 nm in CECL and therefore the dimensions are in many cases nearly optimal. It can be also added that since the hot electrons dilute in rather large volume compared to the normal electrons they have time to stabilize with water molecules or make hydrated hot electrons which can diffuse before finding proper reacting molecules, e.g. luminescent labels or other intermediates. The real picture of the chemistry in the electrically active layer is, however, much more complex. If the illustrative immunological hybridization assay principle is used, the coating antibodies are often randomly bound in the coating process and relatively few of them can bind antigens or analyte. The process of binding of coating antibody in active mode can be different in different places and hence concentrations of the antigen and secondary labeled antibodies are equally well unequally distributed on the surface. Furthermore, in three dimensions, the theoretical model is also not credible since antibodies have two-binding sites and more complex structures can be formed also horizontally and in three dimensions. It results in that in certain places there are too high concentration of hot electrons and other places too low.

The chemical reactions by hot electrons can be improved with better control of chemical composition or/and more control of hot-electrons distribution in three dimensions. The first case is related with the chemical quality of production which depends on fabrication technology, control of deposition process, purity of ligands etc., i.e. this is related to external factors, not to design of the device itself. The second possibility is related to improvement of the device by adding more functions into it. If the flow of the hot electrons can be controlled in three dimensions, it is possible to find optimal excitation method for each case with a computer program under a feedback from the electrically active layer and/or luminescence signals. The present invention brings a solution to this problem. The average energy of hot electrons can be optimized dynamically. The present disclosure demonstrates that significant improvements in performance of ECL devices can be achieved by the precise dynamic control of the injection of hot electrons into the electrically active layer containing the chemically reacting molecules.

Further definitions to structures and principles:

20175624 prh 01 -04- 2019

The barrier layer is the layer the hot electrons tunnel through. The sensing layer is on top of the barrier layer. Only those hot electrons having high energies transit to the sensing layer through tunneling. Energies of some hot electrons can be smaller than the barrier energy though. However, quantum mechanics allows them tunneling through/across the barrier with some probability. The smaller the energy of a hot electron, the smaller tunneling probability. The sensing layer is the layer where ECL occurs.

In this disclosure when electron tunneling is discussed, the barrier layer is an energy band gap structure. The energy band gap structure term is clearer for a physicist 10 because the energy band gap value is included in the formulas. The barrier layer term is a shorter one meaning a barrier for electrons to travel into the sensing layer in a more general sense. The last term is easier for engineers.

Sensing layer is the layer that reaches up to 500 nm above the barrier layer wherein hot electrons produce specific chemical reactions. It is to be understood that in this 15 description the device and its elements are referred to in position as shown in the figures, however a skilled artisan understands that the device may be in any position and thus the sensing layer does not need to be above the barrier layer but it may as well be underneath the barrier layer if the device is turned into another position.

In this disclosure by a layer it is meant a real structure made of a known material, characterized with a precise thickness. However, in regards with sensing layer, its thickness is not precise, because it depends on chemical composition of aqueous solution and penetration depth of hot electrons and ECL process. Depending on conditions, the thickness can slightly vary. So, the thickness of the sensing layer is a variable value.

There is an interface between the barrier layer and the sensing layer. The interface can cause additional effects: it contributes into the total impedance, and, accordingly can affect the tunneling process. The interface can trap carriers/electric charges causing additional local electric potential.

There is a second interface within the aqueous solution. This interface is between the 30 upper boundary of the sensing layer and the rest of the aqueous solution above. The upper boundary is the penetration depth of hot electrons. There are no hot electrons above that second interface. However, photons produced during ECL travel through

20175624 prh 01 -04- 2019 this interface. Obviously, some of them can dissipate in the second interface layer. In order to reduce the dissipation effect, as well as increase efficiency of the ECL, a AC voltage is used to generate fluctuations of charges and carriers that can result in resonant electrochemical reactions at which the ECL process occurs fast. This can happen at a certain resonant frequency which depends on materials and geometry. The effect can be described using effective cross section of molecules and ligands. At resonance, the effective cross section is increased resulting in faster reactions with hot electrons.

Advantages of the invention

The present invention provides at least the following advantages:

a) The dynamically controlled and optimized generation of hot electrons according to the present disclosure allows decreasing the size of the device, increasing a number of materials which can be used in the production, using more label compounds, adding more functional parameters, and significantly improving the sensitivity of detection.

In an ideal case, distribution of hot-electrons is one-dimensional only. This allows only obtaining similar signals from different locations of the sense surface. ECL signal can be integrated in order to reduce the noise. Hot-electrons are distributed over the sensing surface radially with the center in the middle of the sensing surface 20 (See Figure 5 a-c). This allows creating a feedback between the output signal and operating DC voltages applied to the device. Concentric electro-magnetic waves are generated in the depletion layer between the semiconductor film and the sensing layer. The number of hot-electrons produced during tunneling process, are controlled by the local depletion layer in the vicinity of the tunneling. To realize the new design, 25 the methods of semiconductor CMOS process are applied to produce structures on one substrate and producing many devices during the same process. This allows to significantly reduce the size of the device because of a much bigger effective tunneling surface.

b) The present invention allows increasing number of labels that can be used in

ECL and allows increase of sensitivity of detection. There are two main mechanisms for charge transfer, Ox-Red and Red-Ox, that are used to create

excited states of ligands. These can be illustrated by reactions of an aromatic
compound, Ar.
Ar —> Ar+ + e-	(1a)
Ar+ + e- Ar(1 or3^	(2a)
5 Ar + e_ —> Ar	(1b)
Ar Art1 or3)i + e’	(2b)

20175624 prh 01 -04- 2019

Where equations (1a) - (2a) correspond to Ox-Red path and (1b) - (2b) to that of Red-Ox. Both mechanisms are working in parallel. Ideally, if they are independent, they both can produce exited states Ar^{(1 or3f}. However, a non - radiative mechanism 10 which returns Ar⁺ and Ar to the ground state Ar without photon emission may also exist. This could happen when ionized Ar⁺ and Ar are located closely to each other forming a coupled state Ar⁺ - Ar . The coupled state prevents transitions (2a) or (2b) and can lead by electron transfer from Ar to Ar⁺ to two ground states Ar + Ar. These ligands are not connected to each other then. To prevent this, in the present design 15 we apply AC voltage creating a variable electromagnetic field that goes through the oxide to the interface with sensing layer. This variable electromotoric field separates Ar⁺ and Ar enhancing photon emission. The conditions of the example with Ar apply to chelates and ligands, as well. The present disclosure, therefore, allows reducing the non-radiative transitions and increase the sensitivity of the detection.

Similar situation as in the preceding paragraph is observed in the near-infrared photon emission. In the case of Yb(lll) chelates, for example, Yb(lll)-2 where 2 means the heptadentate ligand 2,6 - bis [N, N-bis (carboxymethy)aminomethy]-4benzoylphenol, using cathodic pulsed-polarized oxide-covered aluminum. This chelate is used to generate emission at 980 nm through radiative transition ²Fs/2 —> 25 ²F7/2 of Yb(lll) and a number of different ligands can be used in the process. For example, with the help of peroxodisulphate ions hydrated electrons e’_aq form sulphate radials δΟΛ aq + S2O8²' —> SO4*' + SO4²' (3)

The Ox-Red pathway occurs in accordance with

YbL + SO₄- YbL^,+ + δθ₄ ²’ (4)

20175624 prh 01 -04- 2019

Ybl_^,+ + e’_aq YbL*(5)

These reactions result in formation of exited singlet or triplet states of YbL* that produce a photonic emission. On the other hand, similar to the aromatic compound (Ar), the red-ox pathway is also possible:

YbL + e’aq YbL'(6)

YbL' + SO₄- YbL* + SO₄ ²·(7)

Again, both reactions are possible if YbL^,+ and YbL are separated. However, if these differently ionized chelates are close to each other (YbL^,+ and YbL) and become coupled, both ox-red and red-ox are delayed or even not possible if these chelates transfer to the ground state without photon emission or non-radiative transition. This is related with formation of heavy excitons.

The variable electro-magnetic field can lead to formation of standing waves comprising electrical carries that participate in the ECL process. Particularly, this field may separate excitons following with radiative transitions. One separated or split exciton can produce two photons. Hence, ECL is enhanced.

In the new design of the present disclosure, two components of the tunneling current are exploited, the first one occurs through the oxide (semiconductor) layer without penetrating the sensing layer, and the second component through the oxide (semiconductor) layer through the interface with the aqueous layer. The first component depends on the anodic potential and distance between the electrodes only, the second component produces hot electrons in sensing layer which then participate in creation of exited states of the label molecules. By using feedback mechanisms, both the tunneling processes or only one of them are controlled and a suitable working point with optimal concentrations of exited chelates is found. It is sometimes convenient to exploit the ratio between the two components of the tunneling current avoiding calibration of the measurements.

A tunneling current It is described by the equation

(11) where energies E_gn and E_gp are referred to the Fermi level from the edges of the valence and conductance bands. For intrinsic semiconductor Eq. (11) will have the form ( v Ί

I=/7sinh ξ——

F /2 l ‘p) (12)

A derivative of the tunnelling current can be expressed by the Bardeen formula (C. J. Chen, Introduction to Scanning Tunneling Microscopy (Oxford University Press, New York, 1993)).

—- oc p_s(E_F - eV) dV ' ^F ’ (13) where ps is the density of energy states of a material. The energy gap function is given by (dHdV\ (di / dV) (14)

The derivative (dl/dV)o corresponds to the energy Ef and the energy E is referred to this level. For n- or p-type semiconductors, g(E) is given by

Λ E Ί ξ — +exp

[ E Ί -ξ — F 7 sp 7

(15)

For intrinsic semiconductor

20175624 prh 01 -04- 2019

(16)

The band gap functions calculated for E_g=1 eV, 2eV and 3 eV are shown in Figure 10. It can be seen that the band gap function for E_g=1 eV band gap has the narrowest peak downwards. The E_g= 2 eV band gap corresponds to the curve inbetween the other two, and the third widest curve corresponds to E_g=3 eV. The bandgap value can be measured or calculated from the energy difference between two values of the band-gap function equaling to 0.5. Experimental data are obtained from the measuring tunneling current - voltage characteristics in accordance with

20175624 prh 01 -04- 2019 equations (11) and (12). An ultrathin S1O2 film having thickness of few nanometers can be used as a barrier layer of the device. The precise thickness is defined depending on what type of chelate is used. The thinner the layer the smaller the effective band gap value. Calibration curves of films having different thicknesses can be used to controlling thickness of ultrathin layers to be produced in manufacturing processes.

The design of the present disclosure improves functionality of the device because it avoids breakdown of the oxide due to generation of defects in the oxide film. The defects create traps for electrons and interface states. Usually, the defects are caused by internal stress. To avoid this, the initial film has to be of high quality. It is possible to create high quality oxide or other barrier films using advantages of semiconductor CMOS process to fabricate the whole device.

An electron energy dissipation model by Me Pherson explains degradation of oxide or other barrier films due to injection of electric carriers through the barrier film.

Another explanation of the degradation is based on the energy dissipation model. Both models especially work for low quality oxides. The defects are forming randomly and their number increases with time. When density of the defects reaches a critical value, the breakdown occurs. In the device of the present disclosure the degradation of the oxide has only a low probability.

It is important that the tunneling process is fast and hot electrons participate in creation of excited chelates during relatively short times. During long times hot electrons can be trapped by impurities, surface states, or losing energy in electromagnetic processes. The advantages of nanotechnology are exploited allowing fabrication of small devices. By incorporating the new design and advantages of up25 to-date fabrication processes, small electrochemical devices can be made. As a result, appropriately short times for tunneling process and excited chelates are acquired, and accordingly, significant reduction of number of dissipated hot electrons are gained.

Quenching of electroluminescence is dependent on number of excitons caused by coupling of differently charged chelates and ions. In order to provide an optimal tunneling through the layer, a precise cathodic pulsed excitation should have a precise amplitude of the electro-magnetic pulses to avoid dissipation of electrons in

20175624 prh 01 -04- 2019 the sensing layer, precise duration of the pulses to avoid heating of the surface avoiding a thermal drift, and accurate measurements of the photon energies, signal integration and feedback. This is provided by the electron circuitry shown in Figure 6.

The devices of this disclosure comprise several thin layer structures that provide electrical connections, band-gap barrier structure, sensing layer and a channel or hole(s) on top of the sensing layer for adding samples. The first conductor (1) provides AC voltage through the barrier layer. The second conductor (2) is used for cathodic pulse excitation of the sensing layer. The sensing layer yields ECL, provides interaction of the exited ligands with the sample particles, and reading the tunneling current through the barrier layer. The third conductive layer (3) is for the ground connection. The working function comprises of monitoring spectrum of ECL and the value of the tunneling current. Specific materials of the band-gap structure i.e. barrier layer and the sensing layer allow adjustment to the optical spectrum range. The optical spectrum range is defined by composition of the ligands. The tunneling process in the barrier band-gap structure i.e. in barrier layer is triggered by the cathodic pulse excitation. As a result, hot electrons are formed in the sensing layer participating in chemical reactions. The sensing layer is the layer with excess of hotelectrons in the vicinity of ligands providing ECL reactions. The electrodes above the sensing layer provide measurement of the tunneling current.

The device of the present disclosure is working in its dynamic regime with generation of hot electrons, tunneling and excitation processes, light emission and interaction of hot electrons with the reactive compounds. Sufficient sensitivity is achieved through a large number of excited ligands in the sensing layer providing immediate interaction with the reactive compounds. The tunneling process through the band-gap structure

i.e. barrier layer is described by semiconductor physics. Therefore, there is a close incorporation of physical and chemical processes. Moreover, feedbacks are used to provide control of the chemical reactions. Additional control of the tunneling current through the ratio between the dark tunneling current (tunneling current beneath the electrically active layer) and the tunneling current in the electrically active layer provide means to monitor the reaction process. Hence, both parameters, the luminescence intensity and tunneling current generate cumulative data about the sensing layer.

20175624 prh 01 -04- 2019

All ECL devices use excitable labels of different types. Usually, there are two possible pathways, ox-red and red-ox, that lead to production of excited labels. However, the number of useful labels depends on the non-radiative transitions. The more non-radiative transitions, the less number of excited ligands are found. The non-radiative transitions use coupling of differently charged labels from ox-red and red-ox reactions. An effective way to reduce the non-radiative transitions is to separate these labels and reduce the probability of coupling. As they are charged compounds, variable electromagnetic field can be applied. This is realized by applying AC voltage and creating an interface dynamic layer. The variable electromagnetic field adds oscillatory movement of the charged labels and reduces probability of coupling oppositely charged labels.

Surface plasmon resonance (SPR) Kretschman configuration, nanoparticle plasmon resonance as well as nanoparticle plasmon resonance can be involved into the photonic emission. Some metals like silver, copper, gold, aluminum, platinum can exhibit plasmonic emission. The interaction between plasmons and photoemission can be involved in ECL, especially close to the metal contacts that are contacting with aqueous solution. Characteristics of plasmons and surface plasmons depend on quantum electrical metal properties including dynamics of free electron gas and dielectric function of the free electron gas. Plasmon’s quantum energies are much higher than photon energies of chemiluminesence, thus plasmons can impact on the photonic spectrum. Also non-radiative surface plasmons can couple with photonic emission. These are additional factors that affect the photonic emission spectrum. However, plasmons do not dominate at long distances from the contacts on the surface. US9075018B2 (C.D. Geddes and M. Previte) indicates that metallic surfaces in bioassay systems can enhance ECL-based reactions where metallic surface plasmons are excited by a chemically induced electronically excited state of chemiluminescent species However, plasmons are sensitive to internal electromagnetic field especially in the case of pulsed excitation. The plasmons are not stable, their quantum number can change during photonic emission and, therefore, they can add uncontrolled change into the spectrum. Moreover, any defects and impurities as well as degradation of the metal layer change the electro-dynamics of plasmons. Therefore, instead of using plasmons in ECL process, we limited their influence by choosing materials for sensing layer that are less plasmonic-active.

20175624 prh 01 -04- 2019

Moreover, a relatively large sensing surface with surface with hexagonal geometry was chosen allowing a photodetector to be focused on the central area of the sensing layer eliminating photoemission near edges of metallic components.

There is also a lower energy electronic excitation at metal surfaces, so-called acoustic surface plasmons, that can impact on ECL through phonon interaction. Lowenergy collective excitation exhibits linear dispersion at low wave vectors and might therefore affect the phonon dynamics near the Fermi level (Ritchie R H 1973 Surf. Sci. 34 p. 1, Feibelman P J 1982 Prog. Surf. Sci. 12 p. 287, Liebsch A 1997 Electronic Excitations at Metal Surfaces (New York: Plenum).

The photonic emission spectrum is measured by a photodetector placed at close distance above the sensing area (area from which the ECL is measured). The photonic spectrum is integrated during the decay time. The method is different from traditional measurements because the integration is not started exactly with the excitation time but starting with a relatively short delay. This allows to eliminate spectrum of high energy photons. These photons do not comprise information about molecules in a test sample because the interaction time is longer than the quick photonic response just after the excitation. Moreover, amplitude of the photonic signal from high energetic photons is much higher than for less energetic photons. Therefore, the spectrum of the high-energetic photons will dominate in the integrated signal. The most informative photonic spectrum comes during interaction between the solution phase and molecules of the sample that occurs at longer times. Therefore, the optical spectrum is integrated after a delay from the excitation pulse avoiding non-informative optical signals due very fast interaction processes like plasmon-interactions.

The invention is now described with a set of non- limiting examples:

EXAMPLES

Example 1: Construction of device.

The device as shown in Figures 2 and 3 is fabricated with semiconductor processes like CMOS (Complementary metal-oxide-semiconductor) or by SoC (system on a chip) method. These methods comprise silicon wafer processing in clean room facilities including thin films deposition, lithography, etching, thermal

20175624 prh 01 -04- 2019 treatment, doping, wire bonding and packaging. Substrate 69 can be made of, for example, of p-type silicon. The electrode contacts 62, 63 and 64 can be made with optical lithography and plasma etching followed by deposition of contact material, for example, n-type silicon or a metal. Metal can be deposited by evaporation in vacuum or magnetron sputtering followed by polishing. A resistive or barrier layer is the layer which defines the energy barrier allowing transfer of only those hot electrons having sufficient high energies into sensing layer. The resistive barrier layer is exemplified by a silicon oxide film having thickness of 1-5 nm, preferably 1-3 nm, most preferably 2-2.5 nm. An anode electrode is separated from the resistive layer by a gap and is attached to the substrate with an anchor made of substrate material. There can be more than one anchor. The anode electrode can be made of a polysilicon or transparent conductor. The sensing layer, for example, electrolyte, is placed in between the resistive layer and the anode electrode. Thickness of the sensing layer may be up to 500 nm.

The conductive regions are electrically isolated by an isolation layer. Also there may be one or more holes in the upper structure of the device for observation and detection of photon radiation during ECL reactions.

Example 2: Device with several thin layer structures to provide multiple electrical connections.

A device according to this invention may comprise several thin layer structures to provide multiple electrical connections. FIG. 4 illustrates such a device involving several thin layer structures that provide multiple electrical connections 101, 102, 103, 105, 106, 107, resistive barrier layer 108, and a liquid sensing layer 115. Next 25 to the sensing layer there is one or more channels that could be used to add the liquid to create the sensing layer but also for adding liquid samples into the sensing layer. Contact 107 provides AC (order of couple of volts, within the range 1-20 V) or DC+AC voltage (DC voltage 1-50V) through the band-gap barrier layer and sensing layer. Contact 106 is for cathodic pulse excitation of hot electrons into the sensing 30 layer. The frequency of the pulses depends on the geometry of the device; generally the frequency is in the range of μΗζ. Sensing layer 115 provides ECL after electrical excitation. Sensing layer also provides a reaction milieu for ligands to be excited and

20175624 prh 01 -04- 2019 place for read the tunneling current through the barrier layer. Features 110, 111 in figure 4 are not mechanical structures but indications of electron charge density near surfaces which impact the tunneling hot electrons. Higher electric field allows higher probabilities for electron tunneling process. Figure 4 shows schematically how the top electrode 102 can be deformed due to electrostatic forces. Even very tiny deformation can strongly increase probability of electron tunneling. With no electrostatic forces (no voltages), the top electrode 102 is straight in the horizontal plane.

Example 3: ECL device comprising one cathode electrode for electrical pulse excitation, multitude of reading electrodes around the cathode electrode separated by the tunneling region and a central electrode for exciting variable electromagnetic field.

Figure 7 shows an ECL device comprising a cathode electrode for electrical pulsed excitation, a multitude of electrodes around the cathode separated by the tunneling region, and a central electrode for exiting variable electro-magnetic field. In a simplest case, AC voltage is applied to the central electrode. The shape of the voltage curve can have a more complex form including amplitude modulation, sweeping frequency, or other forms described by mathematical functions. The central region is reserved for electrically active layer that is usually a complex chemical composition with defined luminescence labels or other reacting molecules. Because there are several reading electrodes, the tunneling process can be controlled locally. The tunneling current is integrated and it compares the tunneling currents at different locations.

Several devices of Figure 7 can be manufactured on one substrate. The devices can 25 share some of reading electrodes. An example of such design is shown in Figure 8.

The devices are placed around central axis. This is made for practical applications wherein the ECL can be observed from different sensors. The photonic system is shifted from one sensor to another by rotating around the axis. One of the problems of existing ECL devices is related to the mechanism delivering liquids onto the sensing area of the device. One solution is to keep an aqueous solution in a separate hermetic container, which is mechanically broken before its use. This manipulation is made by an operator during testing. This is naturally inconvenient because it requires

20175624 prh 01 -04- 2019 manual operations by the operator to be part of testing procedure. For an improvement of this inconveniency, in the present design, a micro-container in the device is connected to a micro-pump. The micro-container delivers a desired amount of the measuring solution onto the sensing layer via a micro-channel. The device can also include a microvalve with an electronic control. The advantages are less amount of the solution, and precise amount of it resulting in a uniform film on the surface. The device can comprise a bigger tank and the buffer tank connected with a microchannel with control by micropumps. Such scheme allows having enough aqueous material for multiple tests. The design provides also a large tunneling area which makes ECL more economical for chemicals. Ultrasound waves can be applied to create a more uniform liquid film. The device can be made on one substrate, for example, on a silicon wafer, by semiconductor fabrication processes like CMOS.

Example 4: Preferable materials for making the device.

The barrier layer can be produced with an insulating or semiconductor materials band gap barrier of which provide the tunneling of electrons into the sensing layer after cathodic pulse excitation with formation of hydrated hot electrons capable of interacting with labels in the sensing layer. One of the materials suitable for the sensing layer is an ultrathin silicon oxide having thickness from 1 to 50 nanometers. Preferably the thickness is 1-20 nanometers, most preferably 1-10 nanometers. The silicon oxide film can be prepared by thermal oxidation of crystalline silicon as well as chemical vapor deposition. Other resistor films, including organic ones are described in the literature.

An advantageous pair of materials that can be used is aluminum oxide and aluminum. The layer of aluminum oxide can be formed spontaneously in air on aluminum or can be done in specific conditions as known from textbooks and elsewhere in the prior art. Aluminum oxide can be created also on another base metal or an electrical conductor. Aluminum oxide provides good electrical insulation even at high-temperature, excellent dielectric properties from DC to GHz frequencies. Thick oxide can sustain high voltages. By fabricating films of precise thickness, one can make sensing layer of predicted value of the band gap in accordance with equation (16). Purity of the material is important in order to provide sufficient amount of hot electrons in the conduction band. Impurities in the film trap the tunneling

20175624 prh 01 -04- 2019 electrons causing heating of the sample. Purity of aluminum substrate shall be not less than 99%. The high purity aluminum resists attack by most of gases except wet fluorine. Also it is resistant against many reagents except hydrofluoric acid or phosphoric acid. Those chemicals are very reactive but are not used in diagnostic measurements. A nano-porous aluminum oxide can be also applied. It provides higher electrical resistivity for thinner films. Hence, one can get thinner films keeping the desired band gap value by a proper porosity. Porous films can be created in weakly soluble electrolytes such as sulfuric, phosphoric, chromic and oxalic acid. Particularly, nano-porous alumina films consisting of aluminum oxide films comprising nanosized cylindrical pores arranged parallel to each other in a quasi-hexagonal system. Such films provide tunneling channels across the film and nano-pores prevent tunneling with longer paths in which trapping of electrons is more probable. The nano-porous aluminum oxide films can be fabricated by electrochemical etching of aluminum film in slightly acidic conditions, typically with pH between 3 and 5 but rather closer to 5. The process depends on voltage and current density, anodization time and temperature. Possible solutions for electrochemical etching are sulfuric acid, phosphoric acid, oxalic acid, and chromic acid. The pore size will be different depending on the solution, but, all types of solution provide high uniformity of the pores. Pore size of anodization in phosphoric acid is larger than that of anodization in oxalic acid pore while size of anodization in oxalic acid is bigger than that of anodization in sulfuric acid.

The fabrication process can include deposition of aluminum film by magnetron sputtering with DC or AC, at temperature of around 450°C where the magnetron’s target is aluminum of very high purity, of 99.999%, in an argon plasma at gas pressure less than 3 mTorr. Uniformity of the film depends on size of the magnetron’s target and distance between the substrate and the target. Bigger targets and bigger distances provide better uniformity. After the aluminum film is deposited, the substrate is processed with anodic electrochemically etching.

The silicon oxide film can be fabricated first followed by alumina fabrication process so that there is a stop layer for electrochemical etching providing precise thickness. Other materials can be used as a stop layer too, for example, titanium nitride or porous titanium nitride having nanopores of size of 1 nm.

20175624 prh 01 -04- 2019

Example 5: ECL device integrated to an optical detection device.

FIG. 9a, b illustrates an ECL device integrated to an optical system. The ECL device is assembled on a base, for example, having a cylindrical shape, on top of which a second part is assembled having means to rotating around vertical axis and a window for mounting an optical detection device, for example, a photo-multiplier tube. Position of the optical device can be adjusted over active region of any of the devices enabling measurements of multiple devices. Fig. 9A is a top view of the device and Fig. 9b is a cross section in vertical plane. The device comprises several sensors integrated on a micro-chip, micro-pumps, tanks for liquids, supplied with micro-valves 10 buffer tanks for liquids and other components supplying power and electronic circuit.

The chip is placed in a holder fixed in the cylinder. The cylinder is mounted on a base which is provided by electrical connectors, bus for electronics and data transfer and other cables. The device is mounted in cylinder covered by flange which can be rotated around vertical axis. There is an optical detection device, which is photomultiplier tube, CCD or other related unit, mounted on the flange.

Example 6: Dynamic feedback of hot electron flow.

This invention provides a dynamic feedback of hot electron flow as is shown in Figure

6. The feedback effect works as following, for example. Suppose the number of hot electrons is directly proportional to the input power. The number of excited molecules 20 increases to some extent. The number of excited molecules is limited to the total number of molecules. Accordingly, the ECL intensity increases to some extent too. Here we have coupled systems: hot electrons, excited molecules and photons. The increase of ECL intensity leads to decrease the number of excited molecules. So, there are two processes affecting the number of excited molecules: a) hot electrons 25 produce new excited molecules, and b) radiated photons reduce the number of excited molecules. If the number of excited molecules increases despite the effect b), the ECL intensity continues to increase until both processes a) and b) compensate each other. After some point, the ECL intensity starts to decrease despite further increase of the number of hot electrons. So, we can distinguish a non-balanced 30 process when dN/dn>0 and dECL/dn>0, the balanced process when dN/dn=0 and dECL/dn=0, and again a non-balanced process when dN/dn<0 and dECL/dn<0. Here dN/dn is the derivative of the number of excited molecules with respect to the number

20175624 prh 01 -04- 2019 of hot electrons, dECL/dn is the derivative of the ECL intensity with respect to the number of hot electrons. The feedback mechanism calculates required power parameters to find the balanced regime when the flux of hot electrons is homogeneous and the ECL is stable.

Example 7: Application of the hot electrons to measure luminescence.

Thus far commercially the most important ECL applications is the technique in analytical chemistry utilizing the derivatives of Ru(bpy)-chelates as the source of light generated by electricity. Such labels are detected on anode in a micellar water solution. The technique of cathodic electrochemistry (CECL) has several advantages 10 over the anodic ECL, including independence of micellar solutions and aqueous buffers can be used. Therefore the examples of the present disclosure focus to CECL. It has been previously demonstrated that CECL bioaffinity assays such as immunoassays and nucleic acid probing assays can be successfully carried out on oxide coated silicon a and alumina electrodes with CECL (see e.g. T. Ala-Kleme et 15 al., US7,513,983 and 8,328,968). However, these electrodes have poor stability of the biocoatings during the excitation due to unability to control the excitation process.

The measuring device described in Figure 3 was constructed. The base material (51) was a doped silicon substrate. The substrate was fabricated with CMOS or SoC methods and as described in texts of Figs 2 and 3. The resistive barrier layer 55 was 20 made on cathode, which in this case was silicon oxide of thickness of 3-5 nm. Also various composite materials of graphite and other conductor pastes can be used as electrode materials as known from the prior art.

The sample cavity (56 in Figures 2, 3) was filled with 1.5 microliter of Tb- 2,6-bisN,N-(carboxymethyl-4-bentsoylphenol) 10 ’¹⁰ M chelate solution. In standard experiment sixty pulses were applied with 20 Hz, excitation pulse -22 V, 0.2 ms, delay 0.05ms. When the electrons were accelerated in the electric field before deflection, the luminescence intensity increased along with the increase of energy to electrons. This was according to the calculated increase of volume of excited chelate molecules and diffusion control of the chelate molecules.

The sandwich immunoassay of the C-reactive protein was carried out. The cavity was filled with 50mM Trizma base, 0.05% NaN3, 0.9% NaCI, 0.1% Bovine serum albumin,

6% sorbitol, 1 mM CaCL, pH7.8, containing anti-CRP antibody 7.0 mg/ml (Medix Biochemica Oy, clone 6405). After 2 h the cavity was aspirated and dried at 30°C. The cavity was treated again with one microliter of 10% trehalose solution and dried again. The secondary antibody (Medix Biochemica Oy, clone 6404) was labelled with Tb- 2,6-bis-N,N-(carboxymethyl-4-bentsoylphenol) chelate to get solution of labelled antibody (0.074 mg/ml) in the same buffer as the coating antibody. One microliter of that antibody was applied into the sample cavity. The liquid was dried as above and then sealed air-tightly. The immunoassay was started with addition of 1.5 microliter of CRP standard solution of 10 ng/ml in the same buffer as above. After 5 minutes the reaction cavity was aspirated and filled with 50mM Trizma base, 0.05% NaN3, 0.9% NaCI, 0.1% Bovine serum albumin, 6% sorbitol, 1 mM CaCL, pH7.8. Time-resolved luminescence was measured with using different means of control and feedback systems described. If measurement was used without feedback control with using 60 pulses, 20 Hz, excitation pulse -20 V, 0.2 ms, delay 0.05ms, less luminescence signal was obtained, proving that the feedback control enables to get higher signal output from an analytical assay.

References

20175624 prh 01 -04- 2019

US 5,318,952 Hato

US 6,327,187 B1 Bergemont et al.

US 9,368,209 B2 /2016 Miiani et al.

US 5,716,842 A 2/1998 Baier et al.

US 5,720,922 A 2/1998 Ghaed et al.

US 5,744,367 A 4/1998 Talley et al. US 5,746,974 A 5/1998 Massey et al. US 5,779,976 A 7/1998 Leland et al. US 5,786,141 A 7/1998 Bard et al.

US 5,792,621 A 8/1998 Verostko et al.

US 5,798,083 A 8/1998 Massey et al.

US 5,833,925 A 11/1998 Shu et al.

US 5,935,779 A 8/1999 Massey et al.

US 5,965,452 A 10/1999 Kovacs

US 5,993,740 A 11/1999 Niiyama et al.

US 6,036,840 A 3/2000 Christensen

US 6,066,448 A 5/2000 Wohlstadter et al.

US 6,078,782 A 6/2000 Leland et al.

US 6,133,043 A 10/2000 Talley et al.

US 6,136,268 A 10/2000 Ala-Kleme et al. US 6,140,045 A 10/2000 Wohlstadter et al. US 6,200,531 B1 3/2001 Liljestrand et al.

US 6,207,369 B1 3/2001 Wohlstadter et al. US 6,316,607 B1 11/2001 Masseyetai. US 6,325,973 B1 12/2001 Leland et al. US 6,479,233 B1 11/2002 Bard et al. US 6,517,777 B2 2/2003 Liljestrand et al.

US 6,808,939 B2 10/2004 Sigal et al.

US 8,211,279 B2 7/2012 Bard et al.

US 8,920,718 B2 12/2014 Kulmala et al. US 9,075,018 B2 7/2015 Geddes et al.

US 9,139,766 B2 9/2015 Qian et al.

US 2012/021443 A1 Geddes et al.

US 2014/072963 A1 Qin

US 2014/322538 A1 Bard et al.

US 2015/330986 A1 Grote

US 2016/097765 A1 Kraus et al.

US 2016/131588 A1 Natrajan et al.

WO 2014/131046 A1 Von Borstel et al.

20175624 prh 01 -04- 2019

20175624 prh 08 -02- 2019

Claims (14)

An apparatus for providing a homogeneous flow of hot electrons in an aqueous solution comprising: a cathode electrode (52, 106) on the surface of the substrate (51) and at least a first and a second control electrode (53, 54, 101, 107); a barrier layer (55, 108) separating the anode electrode (57, 102) from the substrate (51), the barrier layer (55, 108) being formed on the cathode electrode (52, 106); an insulating layer (61) separating the control electrodes (53, 54, 101, 107) from the anode electrode (57, 102); and A water-based sensor layer (56, 115) on top of the barrier layer (55, 108); wherein in the device all said electrodes (52, 53, 54, 57, 101, 102, 106, 107) are connected to an electronic circuit, and the control electrodes (53, 54, 101, 107) provide a lateral electrostatic field 15 and an anode electrode (57, 102) providing a vertical electromagnetic field across the barrier layer (55, 108); wherein the device generates hot electrons by electrical cathode pulses, and the hot electrons receive kinetic energy by moving sideways in the lateral electrostatic field between the first and second guide electrodes (53, 54, 101, 107), and the lateral movement changes from vertical motion to vertical motion, where hot electrons with sufficient energy move through the barrier layer (55, 108) through a quantum tunnel, and a homogeneous flow of hot electrons is generated in the aqueous sensor layer (56, 115). Device according to Claim 1, characterized in that the barrier layer (55, 108) is an oxide layer, preferably a silicon oxide or alumina film having a thickness of 1-50 nm, preferably 1-20 nm, most preferably 1-5 nm. Device according to Claim 1 or 2, characterized in that the distance between the anode and cathode electrodes (52, 57, 102, 106) is 50-500 nm. Device according to one of Claims 1 to 3, characterized in that the cathode 35 and the anode electrodes (52, 57, 102, 106) are made on different layers having 20175624 prh 08 -02-20199 provides separate electrical connections and allows for a wide range of geometric designs. Device according to one of Claims 1 to 4, characterized in that The 5-layer (56, 115) comprises ligands which can be excited by hot electrons and which are capable of forming a photon flux after excitation. Device according to Claim 5, characterized in that the anode electrode (57, 102) is horizontally overlapping with the cathode electrode (52, 106). 10 in order to detect electromagnetic emission, preferably photon emission. Device according to Claim 5, characterized in that the sensor layer (56, 115) comprises a sample and the ligand forms the analyte in the sample. 15 with the complex. Arrangement for measuring electrochemical luminescence in one or more water samples, characterized in that the arrangement comprises one or more devices according to any one of claims 5 to 7 integrated 20 microchips (161), and an optical detecting device (160) arranged to measure the photon flux from one sensor layer (56, 115) at a time. An arrangement according to claim 8, characterized in that the microchip (161) is disposed in a holder (157) which is symmetrical, preferably 25, and the symmetrical support (158) is rotatable to change the orientation of the microchip (161) relative to the optical detecting device (160). 10. A method for providing a homogeneous flow of hot electrons 30 in an aqueous solution comprising: a) providing a cathode electrode (52, 106) and at least first and second guide electrodes (53, 54, 101, 107) on the surface of the substrate (51); b) providing an anode electrode (57, 102); c) providing a barrier layer (55, 108) over the cathode electrode (52, 106), 35 to separate the dielectrode (57, 102) from the substrate (51) and 20175624 prh 08 -02-20199 an insulating layer (61) for separating the control electrodes (53, 54,101,107) from the anode electrode (57, 102), and an aqueous sensor layer (56, 115) above the barrier layer (55, 108); d) connecting all said electrodes (52, 53, 54, 57, 101, 102, 106, 107) to an electronic circuit; e) providing a lateral electrostatic field between the control electrodes (53, 54, 101, 107) and a vertical electromagnetic field across the barrier layer (55, 108) by means of the anode electrode (57, 102); f) generating electrical cathode pulses at the cathode (52,106) to produce 10 to achieve these, g) allowing the hot electrons to obtain kinetic energy by moving laterally in a lateral electrostatic field between the first and second control electrodes (54, 54, 101, 107); h) changing the lateral movement by means of cathode voltage pulses to the vertical direction; 15, causing hot electrons of sufficient energy to pass through the barrier layer (55, 108) through a quantum tunnel and to provide a homogeneous flow of hot electrons to the aqueous sensor layer (56, 115). 20 Method according to claim 10, characterized in that the frequency of the cathode electrical pulses is predetermined faster than the time required for the transfer of electrons from one control electrode to another. Method according to claim 10 or 11, characterized in that Tunneling of electrons into the sensor layer (56, 115) is used to excite ligands capable of producing luminescence in the excited state. Method according to one of Claims 10 to 12, characterized in that the method has a feedback control which correlates with the photon process and 30 electron process. Method according to claim 13, characterized in that the feedback times are shorter than the extinction time of the electrochemical luminescence.