WO2015114193A1 - Microdosimeter based on 3d semiconductor structures, method for producing said microdosimeter, and use of said microdosimeter - Google Patents

Abstract

The invention relates to a microdosimeter formed by cells forming an array, where each cell contains a radiation-sensitive volume formed over semiconductors by means of three-dimensional processes defined by the components of a p-n junction so as to ensure the accurate delimitation, to a very small number of μm, of a sensitive volume similar to the mean volume of the cell nucleus, of a diameter approximately equal to, or less than, 10 μm, where the substrate where the cell is produced is a semiconductor wafer and the cell has a diameter of between 5 and 150 μm and a depth of between 1 and 300 μm. The invention also relates to different methods for producing different configurations that said microdosimeters may present, and to the use thereof for detecting radiation in different fields including medical and aerospace applications.

Description

 MICRODOSIMETER BASED ON SEMICONDUCTOR 3D STRUCTURES.

MANUFACTURING PROCEDURE OF SUCH MICRODOSIMETER AND USE OF SUCH

MICRODOSIMETER DESCRIPTION

SECTOR OF THE TECHNIQUE AND OBJECT OF THE INVENTION

This invention relates to a microdosimeter formed by cells that form a matrix, where inside each cell there is a radiation sensitive volume, which is manufactured on semiconductors by means of three-dimensional processes that define the components of a PN junction. The microdosimeter may include a signal processing system obtained with the microdosimeter to obtain radiobiological quantities as the equivalent of H dose by an algorithm comprising geometric corrections and equivalent tissue. Furthermore, the invention indicates manufacturing processes of different configurations that these microdosimeters may have, and their use for radiation detection in different fields including medical and aerospace applications.

The main application is the medical and radiological personal protection area, for example to obtain physical parameters with which to derive the dose equivalent in radiation fields such as those generated in nuclear medicine and hadronterapia, aviation, aerospace exploration, nuclear facilities and accelerators of particles, among others.

STATE OF THE TECHNIQUE

Fluctuations in the micrometric scale energy deposit is one of the most relevant topics in radiobiology. While the energy imparted or absorbed on a macroscopic scale can be considered a continuous magnitude, the dose is deposited in practice through multiple individual interactions of the quanta of the incident radiation in the medium that they pass through. On a microscopic scale the energy imparted behaves like a random variable whose distribution of values follows a certain statistical distribution. The biological effect of radiation depends not only on the energy absorbed in a certain tissue but also on the type of incident radiation that we can call radiation quality. The technique for measuring microscopic energy distributions is known as microdosimetry, this is the study of the ionization distribution or the deposit of energy in microscopic volumes (specific energy), especially in biological matter, as well as its relationship in physical / chemical consequences. and / or biological. Such a branch of dosimetry is essential for both radiation therapy and radiation protection and requires adequate instrumentation for quantification.

A microdosimeter is a device or instrument capable of evaluating the dose (energy deposited per unit mass) deposited in small volumes of micrometric scale masses when placed in a radiation field. Examples of these events are, for example, those that take place on a cellular or subcellular scale (eg cell nucleus) when organisms are irradiated, either for reasons of medical treatment (eg hadron therapy) or for being involuntarily subjected to natural radiation (eg cosmic radiation ) or artificial (eg in nuclear reactors, industry, etc.). The data obtained through a microdosimeter allow obtaining information on the characteristics associated with the deposit of energy dependent on the quality of the radiation. These features are usually inaccessible for conventional macroscopic detectors. The microdosimeters can be used to detect radiation / particles, so phototherapy is used in radiotherapy, and in prototypes, carbon protons and ions are detected mainly, although any other charged particle can be detected. The microdosimeters can be used in the medical field of application for the detection of ionizing radiation, but also in the aerospace area as high energy cosmic particle detectors.

The average diameter of mammalian cells is in the range of a few micrometers. For example, in the case of human cells they have a size ranging from 7 μιη, for red blood cells, to 150 μιη for ovules. When a cellular tissue is irradiated, the radiation dose effectiveness factor and dose rate describe the biological efficacy of different types of radiation to estimate cancer risk. Other factors used to assess biological efficiency are the relative biological efficiency (RBE) and the weighting factor (W _r ). It is the distribution of the dose imparted - which depends on the type of particle and its energy - on structures such as the cell nucleus, which contains the DNA, which determines the level of cellular damage and consequently its survival / repair. Depending on the distribution of the ionization created by the radiation, the cellular DNA will be able to repair itself or be inactivated; Thus, for example, in cases of radiation protection, the first thing would be as desired, while in radiotherapy / hadronterapia the second is what is required (on tumor cells, but not on the surrounding healthy tissue).

The microdosimeters allow estimating radiation parameters in real time such as, mainly, those related to the energy deposited by the radiation, but to quantify the radiobiological damage it is necessary to derive some physical parameters from said deposited energy and perform a postprocessing with an electronics additional and specific software. One of the most relevant dosimetric parameters to quantify the effectiveness of radiation is the Linear Energy Transfer (LET) since it defines the local deposit of radiation energy along the path of the charged particles associated with the exposure. to this radiation and is directly correlated with radiobiological effects on the irradiated tissue (through another radiobiological parameter: relative biological efficacy (RBE)). On a micrometric scale, the LET varies when penetrating the tissue / white (understood in this context as 'white' to a compendium of mixed tissue with variable and highly heterogeneous densities, compositions and geometries, eg human tissue) and for some types of irradiation , such as those used in hadronterapia, its value varies considerably throughout its incidence with the target. But it is also that when the irradiated target has a micrometric size, the energy deposit is not deterministic, but stochastic. Then, in microdosimetry it is necessary to use other stochastic variables to quantify experimentally [ICRU Report 36, Microdosimetry, The International Commission on Radiation Units and measurements, 1983], such as:

(i) 'Specific energy ", z, defined as the ratio between the energy deposited by ionizing radiation, ε, detected in the microdosimeter, and the mass of the white volume that passes through, m: z = - [Gy]

 m

 (in a set of events or for event to event)

(ii) Another magnitude derived from ε (but for a single deposit event) is the "linear energ", /, defined as:

 ε

y ⁼ [keV / μΓΤΐ] being / the 'average rope length', ie average length of randomly oriented strings in a convex V volume, which is calculated as [Kellerer A. and Rossi H., The theory of dual radiation action ", Current topics in Radiation Research 8 (1972) 85-158]:

/ = 4— [ _M m] where V is the microscopic volume of the region of interest irradiated and S the area of such volume. Thus, "and" is the magnitude analogous to the LET at micrometric scales. The microdosimeter must be designed in such a way that its radiation-sensitive volume has an equivalent to the average size of the cellular structures to be studied. The absorbed dose in the microdosimeter can be derived from the distribution of linear energy values ("and") within the material that forms the microdosimetric cavities where the energy of the incident radiation is deposited and the characteristics associated with the quality of the radiation. Said material is the so-called "sensitive volume" and is instrumentally a particle / radiation detector, ie a device or part of it that is technologically capable of characterizing a radiation field. However, it is worth noting that such studies can only be carried out with sensitive volume detectors with sizes comparable to biological cells, ie micrometric volumes [Rossi HH, Zaider M., Microdosimetry and its Applications, Springer 1996]. To make estimates of such physical / micro-dosimetric quantities, proportional gas meters have been traditionally used, such as Tissue Equivalent Proportional Chambers (PTSD) or Rossi Chambers [Rossi HH and Rosenzweig W., Measurements of neutron dose as a function of linear energy transfer, Radiology, vol. 64, pp. 404-411, 1955], which allow to determine the energy transferred through pulse height measurements. These devices have some disadvantages, for example (a) they are relatively large (-diameters> 1 cm), (b) they have effects due to the wall containing the gas, (events created on the walls of the PTSD); (c) limited spatial resolution; (d) PTSDs usually need the replacement of the gaseous medium and also their operating conditions are not so versatile to allow their use in all irradiation conditions, for example for in-vivo applications in hadronterapia; (e) require conditions of low irradiation rates; (f) cannot evaluate / simulate a matrix of cells; (g) its measurement is not faithful if the creep of charged particles varies significantly on a scale of length similar to its size. One of the great attractions of using silicon and other semiconductors in microdosimetry is the possibility of producing structures of micrometric dimensions. In addition, these do not require any gas supply system, operate at low voltage, are light and easily portable. Since the 1980s, several studies have proposed and developed semiconductor detectors based on microdosimetry. The semiconductor radiation sensors act as detectors that quantify the ionization produced by the radiation when it affects a radiation sensitive volume of said sensor. The semiconductor microdosimeters proposed since then can be grouped into three types and are based on PN junctions, where a PN junction is a two-component structure formed by the metallurgical junction of two crystals, such as silicon, germanium, etc., one of type P and another type N. To achieve semiconductors of type P or N, these are usually doped with impurities, eg Boron or Phosphorus respectively. If an electrical conductive part (eg aluminum layer in contact with each doped zone) is attached to said semiconductors P, N, two electrodes are created, one positive and one negative, to which a potential difference can be applied ( "polarization") to create an electric field within said PN junction.

(I) McNulty patented, US 5256879, a microdosimeter based on a series of parallel PN planar junctions, such that each PN junction itself constituted a sensitive volume and this is connected to a standard electronic reading circuit (ie preamp, shaping amplifier, ADC, microprocessor, etc).

(II) The Rozenfeld group shows the first silicon microdosimeters with diode series of rectangular parallelepiped structures made of Silicon-on-Insulation (what is known as "SOI" wafers) [Bradley PD. et al., Solid state microdosimetry, Nucí. Inst. And Meth. B 2001; 184: 135-157]. Recently, Rozenfeld patented, WO080074074, a silicon microdosimeter based on series of planar PN junctions by the diffusion or ion implantation microfabrication technique, but in a surface arrangement in which they partially isolate said PN junctions by the external part to said joints, leaving an outgoing (outer) shape of the approximately cylindrical PN joint.

(III) Another design developed in recent years by S. Agosteo is based on the use of a silicon telescope. Although such devices have a sensitive thickness from ~ 2 μιη, they are manufactured on thick substrates, between 300 and 500 μιη, which contributes to the main signal due to the backscatter generated by the radiation incident on such substrates. These authors have recently patented a method for microdosimetry with this design US8183655B2.

 All these designs have tried to solve the same set of technological challenges: (i) the sensitive volume of the detector must have well-known micrometric dimensions,

 (ii) electronic noise must be as small as possible; However, the designs presented above - with very thin sensitive volumes - have high capacities, of the order of nanofarads or higher,

(iii) corrections must be made to reproduce the spectrum of energy deposited under conditions of tissue equivalence since silicon is not equivalent to tissue (TE),

(iv) if the sensitive volume of the detector is approximated to a parallelepiped (planar PN junctions) or to a cylinder (PN cylindrical junctions), geometric corrections must be made to have spherical equivalence, assuming the spherical shape as the closest to the cellular volume,

 (v) a study of the angular response must be made,

 (vi) the effect of distortion of the electric field on the sensitive volume induced by high-LET particles that causes charge collection outside the emptied region (field-funnelling effect) [Hsieh, C.M., Murley, PC. and O'Brien, R. R., A field-funnelling effect on the collection of the alpha-particle-generated carriers in silicon devices, IEEE Electron Device Lett. EDL-2, 103-105 (1981)].

Although these challenges have been addressed in some of the previous designs, their complexity has meant that they have not yet been fully overcome. It is an example that the measurements obtained from the linear energy spectra (and) with these detectors when compared with the data acquired with a PTSD (taken as a reference) still show discrepancies; Some of the causes are: (i) mismatch due to the shape and dimensions of sensitive volumes, i.e. between the sensitive volume of the detector and the cell phone, (ii) lack of charge collection efficiency.

In 1997 Parker presented a new 3D technology alternative to planar technology (used to date in radiation sensors, where it is manufactured on the surface of the substrate), to create columnar electrodes through the semiconductor substrate [Sl Parker et. al., A 3D-proposed new architecture for solid-state radiation detectors, Sensors (Peterborough, NH), 395: 328-343, 1997. 126, 127] [G. Pellegrini et al, "Technology development of 3D detectors for high-energy physics and imaging", Nuclear Instruments and Methods in Physics Research A 487 (2002) 19-26]. This new technology allows the electrodes to be inside the substrate and face in parallel within it and not above and below it limiting the window through which the radiation enters. Works such as WO 2009/071587 A2 have shown how structures that are based on 3D semiconductor structures can be adapted to be used for detection of unloaded particles such as neutrons. In this case, the proposed technology has a matrix of columnar electrodes (containing the P + or N + part) arranged at the vertices of a square in whose center there is another columnar electrode (type N + or P + respectively). This arrangement causes the sensitive volume around said electrodes to have a well defined geometric delimitation (since the electric field created between said electrodes, although radial, does not confine a sensitive volume with symmetric geometry), ie does not form an equivalent to a simple and clear geometric figure and therefore cannot be used for microdosimetry (since a geometric volume similar to cellular structures or substructures, cylindrical or spherical type) is required.

On the other hand, US20100090118 presents devices for aerospace application (pilots and astronauts). These last works show different technologies, although they reproduce, by means of planar technology, externally cylindrical shapes as sensitive volumes (within the substrate N + and P + standard structures are created by diffusion or implantation and then the surrounding silicon area is recorded (removed) with a circular shaped mask, to leave a cylindrical geometric shape around the PN junction.

As can be seen in these documents, it is not easy to achieve a well-defined geometry of the radiation-sensitive volume in such a way as to ensure that a sensitive volume similar to the average volume of the cell nucleus, approximately equal or smaller, is delimited with precision of a few μιη. at 10 μιη in diameter.

It would be desirable to propose other designs that introduce improvements in the technology of existing microdosimeters to perform detectors that develop this technology and allow it to adapt to new fields such as hadronterapia.

EXPLANATION OF THE INVENTION A first aspect of the invention is a microdosimeter comprising a set of cells that form a matrix characterized in that - the substrate where the cell is manufactured is a semiconductor wafer,

 - the cell has a diameter between 5 and 150 μιη and a depth between 1 and 300 μιη,

- the cell comprises a radiation-sensitive volume,

 - the volume-sensitive is defined by the 3D cylindrical engraving of one of the components of at least one PN junction,

for the production of a readable electrical signal.

The components of the PN junction can form two asymmetric electrodes with different thickness and / or shape. The substrate is a semiconductor wafer of the Semiconductor-over-Insulator (SOI) type, comprising at least one of the following materials: Si, Ge, SiC, CdTe, CdZnTe, GaAs, B ₄ C; or is a standard wafer.

The reading of the electrical signal can be of continuous and / or independent type for each of the cells.

The microdosimeter may comprise layers of various equivalent biological tissue materials.

The microdosimeter can comprise a signal processing system that obtains radiobiological quantities by means of an algorithm. The microdosimeter can correlate the variable measured by the microdosimeter, the energy deposited by ionizing radiation, ε, with radiobiological quantities such as the equivalent dose H by an algorithm comprising geometric corrections and equivalent tissue.

The algorithm can comprise the following stages: as measured by the pulse height or load collected in each cell of the microdosimeter matrix as a spectrum or pulse height distribution in each pixel of the semiconductor that acts as a substrate of the matrix that forms said microdosimeter ,

 b assignment of said pulse height of the corresponding energy deposited (ε) by the radiation in each cell,

c calculation of /, "average string length", defined as l = 4- ^V

 s being V the irradiated microscopic volume-sensitive and S the area of such volume,

 _ ε

d. linear energy calculation, and, defined as y— ^~ j ^~ e. calculation of its linear energy distribution, f (y),

F. calculation of the total absorbed dose distribution in the semiconductor as

 wherej is the average frequency of the linear energy, calculated according to:

g. Correction of tissue equivalence via the scale factor using the ratio of braking powers of the tissue and the semiconductor:

 h. Calculation of the estimated dose equivalent by convolving the data from the previous section with the quality factor, Q (LET) or Q (y)

estimación de la eficacia radiobiológica relativa RBE, según: D ^T '- \ Q (y) -yf ( _y dy

estimate of the relative radiobiological efficacy RBE, according to:

RBE = f ^∞ and ² f (y) Q (y) d [l ₀ g (y)]

J— A second aspect of the invention is the manufacturing method of the microdosimeter of the invention comprising the following steps:

(a) preparation of the semiconductor substrate by microelectronic or MEMS processes,

 (b) cylindrical-3D engraving on the semiconductor,

 (c) creation of metallizations on the P + and N + motifs to create the electrodes for reading the electric charge.

The procedure can have two alternative configurations. In the first, in step a) the central component type P + or N + is created by diffusion or implantation and in step b) a cylindrical-3D ring-shaped engraving, centered on the semiconductor pixel, is made and filled with doped polysilicon or with undoped polysilicon that is subsequently doped with a N + or P + type dopant respectively. In the second one in step a) a component type P + or N + is diffused or implanted or a cylindrical-3D-ring engraving is made that is filled with already doped polysilicon or with undoped polysilicon and later this will be doped type P + or N + and in step b) a cylindrical-3D column is formed in the center of the cell, whose internal walls are filled with polysilicon, which is engraved and subsequently doped type N + or P +.

The cylindrical etching process may comprise a reactive beam process by ion beam or a deep etching process by reactive ions (DRIE).

A thinning process of the substrate can be done by chemical etching or reactive ion processes.

A third aspect of the invention is the use of the microdosimeter of the invention for radiation detection in the field of medical application, preferably in hadron therapy. The microdosimeter can be used to perform measurements at the surface level such as on the surface of the skin or at a biological interface, and / or at a certain depth such as under the skin or irradiated tissue.

The microdosimeter can be coupled to a tissue equivalent material that simulates specific biological tissue including water, muscle and / or bone. The microdosimeter can also be used in other fields derived from other radiation environments, such as secondary neutrons generated in therapy, in mixed radiation fields, including radiation protection in personal use and area monitoring, for portable systems, and in the field of application aerospace

DETAILED DESCRIPTION OF THE INVENTION

In this invention, new designs of semiconductor solid-state micro-meters with a cylindrical-3D technology based on 3D-columnar technology of radiation sensor microfabrication are presented. This technology allows the electrodes to be processed inside the wafer instead of being implanted on the surface. Said 3D-columnar technology is based on structures containing perforated column electrodes in the semiconductor substrate [G. Pellegrini, et. al., Fabrication and simulation of novel ultra-thin 3D silicon detectors, NIMA 604 (1-2): 115-118, June 2009]. These columnar structures are preferably cylindrical, but according to the technology used they can be quasicylindrical, prismatic or quasi-spherical.

The design configuration of such microdosimeters is of the pixel type, such that an array of cells (205) is created, and the center of the cell (205) defines the position of the pixel itself. The term "cell" (205) is understood as the simplest portion of the microdosimeter which, when repeated by translation, reproduces the entire assembly, presenting translation symmetry centered on the pixel. The step (108) between two cells (205) is what we define as "piten". The interior of the cell (205) (Figure 2) contains a "radiation sensitive volume" (208) made of a semiconductor material that is capable of detecting the energy deposit left by the eh pairs (203) created by the ionization produced by radiation (201) when passed through it. The microdosimeters can be manufactured to cover from micrometric areas up to cm ² . In addition, said cell matrix (205) can be replicated on a large scale to simulate cell tissues through the ability to "large-scale integration" (VLSI), possible in the process of manufacturing MEMS devices.

The operation of the microdosimeter of the invention is based on the fact that each cell (205) of the microdosimeter acts as a microsensor on a cellular scale (Figure 2). When the radiation or particles (201) pass through the cell (205) of the microdosimeter, it ionizes matter, particularly the sensitive volume (208) of the cells (205), thereby creating charges, in particular, electron-hollow pairs ( eh) (203) in that volume (208). The total number of pairs eh (203) created is proportional to the energy of the radiation or particles (201) that they affect said cell (205). The charges, when they reach the electrodes (formed by the components (104) and (105) of the PN junction and a metal that makes contact in each of them (209) and (210) respectively) give rise to a pulse of current read with a reading electronics (106) gives information of the generated load (203), and therefore of the energy deposited by the radiation (201). Specifically, as the electrodes comprise the components of a PN junction (manufactured in the cell (205), between (104) and (105) or between (602) and (601) in another type of configuration), when metallized (this is to put a conductive material on top of each part of the PN junction that functions as an ohmic contact) and operating in the reverse polarization mode, by applying a potential difference by connecting (206) between the cylindrical electrodes (209) and the central electrode (210), the pairs eh (203) will be separated by the electric field in opposite directions until they reach the electrodes (209 and 210). These pairs eh (203) induce an image charge on the electrodes (209 and 210), which in turn results in a current pulse, which is read in (106). Thus, a first object of the invention is a microdosimeter, hereinafter referred to as a microdosimeter of the invention, formed by a set of cells (205) that are part of a matrix characterized in that

- the substrate where the cell (205) is made is a semiconductor wafer (101), - the cell (205) has a diameter between 5 and 150 μιη and a depth between 1 and 300 μηι,

 - the cell (205) comprises a radiation sensitive volume (208),

 - the sensitive volume (208) is delimited by cylindrical-3D engraving of the components of at least one PN junction,

for the production of a readable electrical signal produced by the charge generated by the radiation.

The components of the PN junction can have different thicknesses and shape giving rise to two asymmetric electrodes. The configuration of asymmetric electrodes by perforating the semiconductor of the substrate (101) that limit the sensitive volume (208) is a configuration that allows that there is no contact layer in the window through which the particles enter, but which at the same time allows there is more sensitive volume (208) with a charge collection close to 100% without the inner intrinsic zone of the column electrode (Example 3). The substrate can be of two types: Semiconductor-over-Insulation (SOI) wafers, where the semiconductor (101) can be Si, Ge, SiC, CdTe, CdZnTe, GaAs, B ₄ C, among others., Eg Figures 1A and 1 B: or standard wafers, eg Figures. 1 C and 1 D. The microdosimeter can be designed with cells (205) of such small areas such that the diameter of the cell (205) is comparable to the average diameter of the biological cells (204), of the order of μιη, and with thicknesses from 1 μιη, such that the sensitive volume (208) is equivalent to that which has, on average, a cell. That is, it is able to approximate its volume to the size of a single cell structure or substructure (Figure 2). However, since the study volume is the cell nucleus 202 (since it contains the DNA, object of study since it is where the radiation will or will not damage said entity and this is what determines whether the irradiated cell will be repaired and replicated or not), the cell (205) in turn contains a smaller volume of equivalent size to the cell nucleus, the so-called volume-sensitive (208), the only volume really sensitive to radiation within the cell (205), of which obtains the energy deposited by the pairs eh (203) of the radiation (201) that crosses the cell.

The sensitive volume (208) is delimited by the cylindrical-3D engraving technique. This three-dimensional etching (or definition or attack or subtraction) process of semiconductor material (101) includes physical or chemical processes of ionic attack or other etching technique. With this technique, part of the substrate is removed creating a three-dimensional perforated shape inside the semiconductor (101) (Figure 2). This perforation leaves a hollow volume in the semiconductor (101), preferably a 3D cylindrical engraving with a column (601) or perforated cylindrical ring (105). When filling that perforation (105) of a semiconductor material for example: polysilicon, which is doped with B or P to obtain the parts of P + and N + junctions respectively, the material that is between the two components, mainly a semiconductor (101) , will be the sensitive volume (208) of the cell (205) (Figure 2). The sensitive volume (208) is a volume delimited by the components of the PN junction. The dimensions of the microdosimeter are controlled by means of the resolution of the equipment for the definition of microelectronic structures, including lithographic processes and therefore allows to clearly define the volume of the cell (205) with dimensions in the area of the micrometers, an essential requirement for the application microdosymmetric In Figure 5 the volume-sensitive (208) is "delimited" by the internal part of the cylindrical-3D engraving (105) and the entire internal part of said cylindrical volume is volume-sensitive (208) of the semiconductor (101) to the radiation, so when a charged particle, for example proton or carbon ion (201) passes through it, it will ionize said medium leaving pairs eh (203) as a function of the initial energy with the I will interact there and these charges will be collected by the electrodes (209 and 210). On the other hand, Figure 6, where a cylindrical-3D-columnar engraving is presented in the middle (601), the volume-sensitive (208) is also intended to be the same as in Figure 5.

The two components of the PN junctions are three-dimensional, and they can have one of the following two configurations.

- cylindrical-3D engraving or cylindrical ring where the sensitive volume is limited at its outer limit by a cylindrical 3D engraving (105) in the form of a ring whose diameter (107) and height (109) are controlled in the manufacturing process. In the case of a ring, the external component of the PN junction is produced by doping the inside of said ring, formed by a perforation in the semiconductor (101) in the form of a cylindrical ring (105), and on whose walls (hollow) deposit polysilicon and then its walls are doped by diffusion P + or N +, or else the polysilicon is already doped; and a central disk (104) where the other component of the PN junction is formed that is formed by diffusing / implanting, in the center of said central shape (104), a dopant to create a N + or P + type motif respectively (104) . (Figure 5).

- cylindrical-3D-column engraving (601) centered on the cell (205), formed by a hollow column (601) whose internal wall is filled with polysilicon, this is recorded in turn and then its walls type P + or N + are doped , which forms the central part of the PN junction. Said columnar engraving will be surrounded by a ring or (i) cylindrical-3D type engraving (as in the previous case, but with a lower height) or (ii) of implantation or diffusion N + or P + (602) respectively. The result is cell (205) of Figure 6.

With these configurations the limitations of previous works are overcome whose planar technology used gave rise to a matrix of cylindrical protruding forms, type P + and N + that are unions by diffusion or implantation, WO080074074. On the other hand in WO 2009071587, the column electrode array has a cell delimitation (205) not entirely cylindrical, while in the case of this invention each cylindrical-3D engraving (105, 601) is either surrounding a P + or N + implantation (104) or surrounded by it (602), respectively. In any case, the cylindrical shape of the volume-sensitive (208) is more clearly defined. The configuration obtained with the cylindrical-3D engravings gives the microdosimeter several advantages over the devices mentioned in the prior art such as the following: the geometry of the cell (205) can be manufactured with a well-defined volume and controlled by the use of Latest generation techniques in MEMS devices and the volume can be equivalent to the average size of the cells that contain the tissue to be irradiated, object of study. The sensitive volume (208) is well confined in a cylindrical region of known dimensions and controlled in the manufacturing process, this sensitive volume (208) can have a diameter (107) between 5 and 150 μιη and depths (109) between 1 and 300 μιη. This covers a wide spectrum of cell types, eg the size of eukaryotic cells of radiobiological interest - animal and human - ranging from 2 to 200 μιη in average diameter, with exceptions. Also, the electric field created by said configurations facilitates the complete emptying of the PN junction so that all pairs eh created by any radiation would be collected entirely (example 3). The pixel configuration of the detector, that is to say that it forms a matrix that covers a controllable surface area (equivalent to a portion of irradiated tissue) allows the radiation traces to be independently delimited since each cell (205) is equivalent to a volume-sensitive (208) cell. In addition, the 3D architecture avoids the 'field-funnelling effect' since it favors the confinement of the load of the pairs eh thanks to the design of well-defined empty volumes. Finally, 3D technology is resistant to radiation because its structure is based on 3D technology for radiation detectors. This advantage makes it especially useful for use in high-rate environments, such as radio / hadron therapy. The proposed technology allows to manufacture the smallest silicon radiation detectors to date (from 2 μιη thick with 3D electrodes), but which can be additionally emptied from the back of the wafer (etch the wafer support silicon SOI ) so that there is a "membrane" (cellular) of 2 μιη thick (avoiding the backscatter of the particles that could interact there). To date, one of the smallest silicon radiation detectors has been made by S. Agosteo using planar technology, although without emptying the wafer, with mechanical support and for them it also acts as a sensitive volume, about 500 μιη thick .

The reading of the unit cells (106) that collect the information from the cells (205) can be continuous or independent type (Figures 7). In the case of continuous reading, the central metallizations (210) are connected to metal pads (701) with a metal track (704) and the external metallizations (209) are joined with a metal track (703) that connects the cylindrical electrodes with strip 702 to polarize the PN junction. In the case of reading independent (Figure 8) each cell (205) is read independently, where individual tracks (802) connect with independent pads (801). The pads are to which the reading electronics (106) will be connected to read the obtained signal. This type of reading has two advantages: on the one hand the connection with the reading electronics via "bump-bonding" [Xiao H., Introduction to Semiconductor Manufacturing Technology, Prentice Hall 2001] is avoided, which could introduce signal by activation, dispersion, ie unwanted contributions to the main signal, it is possible to promote the independent reading of each cell (205) (Figure 8) and avoid the technological complexity of bumping on very thin wafers (which could cause fracture of the same).

It is possible to add additional outer layers in the microdosimeter (and independent of it) of various materials that allow different equivalents of biological tissue and therefore when the radiation passes through these layers (as in the realistic cases of hadronterapia) loses energy and reaches the microdosimeter in a lower layer (as if it were different overlapping fabrics).

Another object of the invention is the method of processing the signal obtained with the microdosimeter in a radiation field in order to correlate the variables measured with radiobiological quantities such as the equivalent of dose H, this is done with an algorithm that contains geometric corrections and of equivalent tissue. Although the device is capable of quantifying the energy imparted (ε) by the incident radiation, it will be in the postprocessing / data analysis where these magnitudes must be convolved to evaluate the parameters that predict radiobiological effects. This procedure can be integrated into a single end device.

The microdosimeter can be coupled to analysis means, such as a Multichannel Analyzer (MCA) to acquire the spectrum in energy for its subsequent post-processing until the dose equivalent (H) is estimated. The analysis means allows to implement an algorithm that includes equivalent tissue correction and geometric correction steps. An example shows a possible algorithm tissue-equivalent correction:

 Tissue Silicon

 D D (Bragg-Gray ratio)

In the previous relationship D. ^'Silicon energy deposited in the silicon microdosimeter, gi ⁱ ssueygb ⁱ nthe _example Powers _{coc¡ente of} stopping, or scale factor, and therefore D ^"ssue is the energy deposited into tissue. Note that the scaling factor it depends on the type of incident particle and its energy [Rossi HH, Zaider M., Microdosimetry and its Applications, Springer 1996], which assumes two conditions: (i) the deposited dose is only due to charged particles and (ii) creep does not change on the irradiated target

(b) geometric correction, will depend on the form to simulate and the geometry of the microdosimeter. this can be done by studying the distributions of the "average string length", /, for the microdosimeters and comparing them with a cylindrical PTSD, using the method proposed by Kellerer [Kellerer, AM, Criteria for the Equivalence of Spherical and Cylindrical Proportional Counters in Microdosimetry, Radiat. Res. 86, (1981) 277-286]. This value is given once the design parameters of the microdosimeter have been set.

These corrections can be made at a later stage or simultaneously with the measurement process itself (with the convenient integrated processing software). More specifically the procedure consists preferably, but not exclusively, of the following steps [Rossi HH, Zaider M., Microdosimetry and its Applications, Springer 1996]: a) measurement of the pulse height (potential difference produced by the charges of the pairs eh (203) when read in 106) in each cell (205) of the microdosimeter matrix, ie spectrum or pulse height distribution in each pixel of the semiconductor that acts as a substrate of the matrix that makes up said microdosimeter.

 b) assignment of said pulse height, after adequate calibration with a reference source or beam, of the corresponding energy deposited (ε) by the radiation in each cell.

c) calculation of /, "average string length", defined as V being the microscopic volume-sensitive (208) irradiated and S the area of such volume. The values of V and S are fixed according to the dimensions of each design, ie diameter (107) and height (109) and therefore is determined geometrically. These values are calculated during the manufacturing process of the microdosimeter

 _ ε

d) calculation of linear energy, and, defined as y— ^~ j ^~ e) calculation of its linear energy distribution, f (y).

f) calculation of the total absorbed dose distribution in the semiconductor as

wherej is the average frequency of the linear energy, calculated according to:

g) correction of tissue equivalence via the scale factor using the ratio of braking powers of the tissue and the semiconductor:

h) calculation of the estimated dose equivalent convolving the data from the previous section with the quality factor, Q (LET) or Q (y) summarized in Table I [ICPR-36]:

H = ^J - = = D ^Tlssue Q {y) ^■ d {y) ^■ dy

and ^J Table I. Quality factor Q (LET)

i) estimate of the relative radiobiological efficacy RBE, according to:

RBE = r _and ² f (y) Q (y) d [log (y)]

 J— Thus, with the final information obtained from the RBE, the radiobiological effects of irradiation at the cellular level can be estimated.

The device can be calibrated with radioactive sources or beams of well-known energy particles so that the dose deposited on the silicon by the incident radiation is directly derived. For this, a standard electronic reading system is coupled so that the voltage pulses induced by the passage of the radiation in the cells are deconvolved to a final equivalent dose reading (H).

As an illustrative example, when a particle or radiation (201) crosses the microdosimeter produces e-h pairs (203) that will ultimately generate a current pulse. Said pulse is sent to the reading electronics (106) which contains a preamplifier that integrates the voltage pulse over time and generates a signal that is proportional to the number of e-h pairs (203) deposited by the radiation in the cell; The reading electronics (106) also contain a shaping amplifier, which amplifies the signal that is then converted into a digital signal via an analog-to-digital converter (ADC); subsequently it is registered to be analyzed by a pulse height analyzer. Thus, thanks to this data processing chain, the amount of energy deposited by the incident radiation in the sensor is inferred.

Another aspect of the invention is the process for manufacturing the microdosimeter of the invention, which comprises the following steps: (a) Preparation of the semiconductor substrate (101) by microelectronic or MEMS processes. This stage is followed depending on the type of cell (205) (Figure 5 and 6):

(a.1) For cell (205) type Figure 5: First, the central component type P + or N + (104) is created by diffusion or implantation.

 (a.2) For cell (205) type Figure 6: either a P + or N + type component is disseminated or implanted or a cylindrical-3D-ring engraving (602) of height (605) is made, which will be filled with already doped polysilicon or with undoped polysilicon and later this will be doped type P + or N +.

(b) Cylindrical-3D engraving on the semiconductor (101).

Depending on the cell (205) that we use (Figures 5 and 6), the following will apply: (b.1) For cell (205), type Figure 5: the cylindrical-3D engraving (105) is performed as a ring, centered on the pixel corresponding to the cell, and then filled with doped polysilicon or undoped polysilicon which is then doped with a N + or P + type dopant respectively (Figure 5). (b.2) For the cell (205) type Figures 6: a cylindrical-3D column (601) is formed in the center of the cell, whose inner walls will also be filled with doped polysilicon or undoped polysilicon that is subsequently doped with dopante type N + or P + respectively (Figure 6). (c) Creation of the metallizations (209) on (105) (in cell type Figures 6) or (602) (in cell type Figures 6), and the other metallization of (210), on the central component (104) or (601) respectively and independently. The union of metallizations with their corresponding P + or N + motifs constitutes the electrodes themselves of the cell. In step (c) the metallizations allow the connection with the reading electronics (106) of the load, pairs e-h (203), which has deposited the radiation or incident particles (201) on the cell (205).

The cylindrical-3D etching process is preferably performed using deep reactive attack techniques with ion beam (Deep Reactive Ion Etching or DRIE). Additionally, part of the wafer insensitive volume can be eliminated by thinning the substrate by chemical etching or reactive ion processes, avoiding possible backscattering or backscattering contributions to the signal during irradiation of the secondary particles generated in the irradiation, thus The energy resolution is improved, as it avoids contributions around the sensitive volume (208), which is the area that will eventually simulate the cell (Figures 3 and 4). This microdosimeter configuration is especially indicated in the case of hadronterapia since the thicknesses between 2 μιη and 10 μιη (lower than the thicknesses of conventional wafers) are suitable for heavy particles used in hadronterapia, eg carbon ions, helium, neon, lithium, silicon, etc. In radiotherapy a low thickness of the detectors (109) would not be very suitable since the detection of the radiotherapy particles (photons) would be very low because the photon absorption is proportional to the volume of the semiconductor (101). Another object of the invention is the use of the microdosimeter of the invention for detection of radiation in the field of medical application, preferably in hadronterapia where it is necessary to detect protons and carbon ions mainly. The microdosimeter can perform both surface level measurements, eg surface from the skin or biological interface, or at a certain depth, eg under the skin or irradiated tissue.

In addition to its direct use in applications of microdosimetry in ex-in-vivo hadronterapia, the microdosimeter of the invention can be coupled to a tissue equivalent material that simulates specific biological tissue, for example water, muscle, bone, etc., in order to that the latter generates the particles - due to the ionization of the primary beam in the tissue-equivalent - that the microdosimeter can in turn detect by means of the method of the invention, to carry out other dosimetric research studies. In this way the fields used in radio / hadron therapy could be parameterized, as well as other fields derived from other radiation environments, e.g. secondary neutrons generated in these therapies. The microdosimeter can be used in other fields derived from other radiation environments, such as secondary neutrons generated in radio / hadron therapy.

Additionally, mixed radiation fields can be included using a matrix of multiple microdosimeters so that each of them is more sensitive to each type of radiation (in a mixed radiation field). This field includes radiation protection in personal use and area monitoring, for portable systems, hadronterapia and in the field of aerospace application, such as detection of high energy charged cosmic particles.

DESCRIPTION FIGURES:

Figure 1. Different geometries and substrates of the microdosimeter of the invention A)

Diagram of the cross section of the microdosimeter manufactured in Wafer type Semiconductor-on-lnsulator (SOI) B) Three-dimensional scheme of A C) Diagram of the cross-section of the microdosimeter manufactured in a standard silicon wafer D) Three-dimensional scheme of C.

Figure 2. Equivalence of the functioning of the biological cell and microdosimeter cell of the invention. A) Biological cell scheme B) Example of cell of the invention C) Diagram of the cross-sectional view of cell B.

Figure 3. Microdosimeter on SOI wafer with thinned substrate

 A) Cross section B) Three-dimensional scheme.

Figure 4. Microdosimeter on silicon wafer with thinned substrate A) Cross section in thinning process B) Cross section already thinned.

Figure 5. Cylindrical-3D-ring engraved type cell. Upper scheme: three-dimensional scheme. Central figure: cross section of said cells with diffusion or central implantation type P + surrounded by cylindrical-3D engraving type N +. Lower scheme: idem than superior but with doping (P + or N +) inverted.

Figure 6. Cylindrical-3D-column engraved type cell: Upper scheme: three-dimensional scheme. Central and lower figure: representation of a cell with cylindrical-3D-column engraving, whose internal walls are doped type P + / N + and is surrounded by an implantation ring N + / P + respectively.

Figure 7. Diagram of the connection of the reading electronics of the microdosimeter with continuous reading Figure 8. Diagram of the connection of the reading electronics of the microdosimeter with independent reading. Figure 9. Field distribution model inside the cell limited by cylindrical ring. A) Electric field distribution B) Potential distribution C) Voltage intensity curves D) Voltage capacity curves. Fig. 10a. Sketch of the arrangement of a microsensor (not to scale) of example 4. The holes of type n and of type p are connected with metal lines.

Fig. 10b. Scheme of two microsensors of example 4 at a distance P. Fig. 11 a. SEM image of the top view of a microdosimeter with 9 μιη diameter, 100 μιη field and 6 μιη thickness, described in example 4.

Fig. 11. SEM image of the top view of a fabricated microsensor equal to that designed in Figure 11 a.

Fig. 12. Tension characteristic curve of some track type microdosimeters in one of the wafers manufactured, described in example 4.

EMBODIMENT OF THE INVENTION

Example 1. This example shows the structure / manufacture of a SOI wafer microdosimeter

The base material is SOI wafers with active silicon type N, high resistivity and thickness (109) between 1 μιη and 100 μιη. After a standard cleaning (RCA or other) a thermal silicon oxide is grown to passivate the silicon surface.

At the top of the wafer, a small circular window, for example 4 μιη in diameter, is opened in silicon oxide by ion-reactive etching or wet etching. The silicon is then implanted or diffused with boron or other P-type dopant to create a P + contact, of the PN junction. The cylindrical-3D ring-shaped engraving (105) (of typical width 3 μιη), and centered on the previous P + contacts, is etched on the active silicon by deep-reactive ion-reactive etching DRIE (with the Bosch process for example) up to the buried rust. Polysilicon is deposited by chemical vapor deposition techniques or other conformal deposition technique. This material is doped with phosphorus or other type N dopant, during deposit or after that, to create an ohmic contact on the walls of the cylindrical-3D ring-shaped engraving. The polysilicon is defined by photolithography and is recorded to form separate contacts around the ring. An intermetal dielectric with a typical thickness of 1 μηι and composed of silicon oxide, silicon nitride or other dielectric suitable for the protection of the holes of the following processes is deposited by vapor phase chemical deposit. In this dielectric, pathways to the P + and N + contacts are recorded by ion-reactive etching or wet etching.

To define the interconnection lines and the pads for the connection, which can be done by ultrasonic spinning (wirebonding) aluminum is deposited by physical vapor phase deposit and is defined by ion-reactive etching or wet etching. Finally, to open the wirebonding pads in aluminum, a passivation, for example a silicon oxide and silicon nitride bilayer which is defined by ion-reactive etching, is deposited by chemical vapor deposition assisted by plasma.

There is a total of 100 unit cells of 80 x 80 μιη ² and the total sensitive area of the microdosimeter is ~ 1 mm ² .

Example 2. Structure of the microdosimeter suitable for hadronterapia

Proceed as in the previous case but the layers of non-sensitive silicon (103) or (405) on the back of the wafer are engraved to form a window whose area corresponds to the active area of the sensor. In the case of silicon wafers, the silicon on the back of the wafer is treated with ion-reactive etching or wet etching until the buried oxide is reached. In this way a thinned substrate is produced.

Irradiation can be performed interchangeably both on the upper face of the microdosimeter and on the lower side. Since (107) and (109) are 10 μιη respectively, the cell is a cylinder with a diameter equal to its height.

Example 3: Structure model of the PN junction manufactured with 3D cylindrical engraving and technological processes that shows its viability as a microdosimeter.

Operating model performed with the "SENTAURUS" semiconductor electrical simulation software (Synopsys TCAD Sentaurus, a finite element semiconductor simulation package, reference: Synopsys TCAD Sentaurus, http://www.svnopsvs.com/tools/tcad/Paqes/default.aspx as of December 12, 2013).

Figures 9A and 9B show the distribution of electric field and potential in a cell. The example of this geometry in both figures refers to a cell as defined in Figure 5, with a diameter (112) of the motif (104) of 20 μιη and a diameter (113) of the motif (105) of 5 μιη and a diameter (107) of the volume-sensitive of 70 μιη. The PN junction designed there (created between (104) and (105)) through the metallizations (209) and (210) give rise to an electric field. In Figures 9 A and 9B it is observed how the emptying of the total sensitive volume (208) is achieved and said sensitive volume (208) has a well defined geometrically cylindrical shape that simulates the cellular structure and / or substructure. When achieving total emptying of the sensitive volume, it is indicated that when passing a charged particle (201), e.g. proton or carbon ion like those used in hadronterapia, all the electron-hollow pairs (eh) (203) that it generates in its path, ionizing the volume-sensitive medium (208), reach the metal contacts (209 and 210) and will produce the signal that is read by the reading electronics (106).

All e-h pairs (203) are clearly collected at the metal contacts (209 and 210) and therefore the charge collection efficiency (CCE) is 100%. That is, all the energy deposited by ionizing radiation, ε, will be totally measured by the volume-sensitive (208). This is an indispensable requirement to directly evaluate the total energy of the incident radiation, since if the CCE is not 100%, the linear energy, ε, is not a reliable value and therefore the derived quantities would not be real. Figures 9C and 9D represent the electrical characterization curves derived from the previous simulations, the intensity-voltage curves IV and capacity-voltage CV respectively. In IV it is seen that each cell (208) thus designed would have its full volume-sensitive with full emptying when 20 V is applied in (206) and that the theoretical inverse current is low

(<1 e ^"12 A), ie good for later operation. Figure 9C shows that at 20 V the capacity for complete emptying is less than 10e ^{" 15} F. These values show that, first this is a device of "low consumption", compared to others cited in the state of the art, such as ionization chambers, which need tens of V to operate and second that the intrinsic noise is very low.

Example 4. Silicon microdosimeters based on cylindrical diodes. A new microdosimeter has been developed as the base detector for microdosimetric applications. These devices are manufactured in three types of SOÍ wafers with a substrate of type n high resistivity and with active volumes of 6, 10 and 20 μιη thickness, for each type of wafer. The collector electrodes are columns recorded through silicon instead of being surface implants as in standard flat diodes, which allows for a much smaller capacity and therefore a lower electronic noise compared to a flat sensor of the same thickness. The sensors are manufactured in cleanroom facilities. Figure 10 shows the schematic arrangement of these microdosimeters showing the pn electrodes and the metal strips that connect them to the contacts: the p-electrode has a diameter of 4 μιη and is surrounded by n-ring holes of 3 μιη thickness with 6, 10 and 20 μιη depth (for each type of wafer) distributed in a square geometry. A wafer contains microdosimeters with locations of 25, 50, 100 and 200 μιη (where P is the distance between p-columns) and with an internal diameter (D) of 9, 10, 15, 20 and 25 μιη, in order to include a greater number of distribution of cells and sizes. The p-type electrode is circular and an ion implantation with boron (p +) has been carried out. A cylindrical ring is etched using the technique of deep etching by reactive ions (DRIE), and then partially filled with phosphorus doped polysilicon (n +) to form the pn junction. The top of the holes is metallized with aluminum and each electrode is connected with an aluminum line to provide electrical contact.

Each microdosimeter consists of 121 independent microsensors arranged in a square matrix. Three main types of detector structures were developed: track detector (pixel-array detector), strip detector and pixel detector. The simplest configuration is the track detector in which all the n + electrodes are connected to the n + contact on one side of the sensor, while all the p + electrodes are connected to the p + contact on the opposite side, thus collecting all the load in all unit cells or sensitive volumes. In the strip configuration, the consecutive p-type electrodes are aligned, resulting in a strip of microsensors connected in a row. In the pixel configuration, although it is usual to use bump-bonded for reading the electronic chip according to the fiip-chip technique, in this design each microsensor is routed through a metal line to a connecting pad for reading electronic microtira to make reading easier. Tapping the wafer silicon support SO! from e! back side on this side, it is expected that finer microsensors will be obtained for the three configurations mentioned above. The Figure 11 shows two SEM images of a processed obiea that contains some pixel microsensors.

The microdosimeters are connected to an appropriate electronic reading system to carry out the experimental tests. The results of the electrical characterization of the first track-type prototypes are detailed below.

An initial function test was carried out on the obiea. The current versus reverse voltage characteristics of the manufactured diodes were obtained with a semi-automatic probe station and an HP4155 Semiconductor Parameter Analyzer. Figure 12 shows the results for some track type microsensors. These pad-type microdosimeters have a reading channel, a total area of 2000 x 2625 im ² and their microsensors are distributed in an 11 x 11 pixel grid. IV curves show a typical diode behavior. The devices are fully functional, as they may be predisposed to reverse voltages higher than complete exhaustion (<1V) and leakage currents are low.

In summary, the innovative microdosimeters described above based on 3D-cylindrical structures with 8, 10 and 20 pm thickness, and variable dimensions and fields, that is, 9, 10, 15, 20 25 m internal diameter and with locations 25, 50, 100 and 200, have been successfully manufactured. These first generation of track-type microdosimeters based on a cylindrical 3D structure optimized to create microsensors shows the feasibility of lowering to the level of the average cell size, and thus providing a closer measurement of silicon AE.

The use of these 3D microdosimeters could improve the accuracy of the RBE calculations normally affected by the inherent uncertainty of Monte Cario simulations, due to the approximation of the material composition and the energy-dependent physical laws involved in these calculations.

Claims

1. Microdosimeter comprising a set of cells that form a matrix characterized in that - the substrate where the cell is manufactured is a semiconductor wafer,  - the cell has a diameter between 5 and 150 μιη and a depth between 1 and 300 μιη, - the cell comprises a radiation-sensitive volume,  - the volume-sensitive is defined by the 3D cylindrical engraving of one of the components of at least one PN junction, -for the production of a readable electrical signal. 2. A microdosimeter according to claim 1 wherein the components of the PN junction form two asymmetric electrodes of different thickness and / or shape. 3. Microdosimeter according to any of claims 1 and 2, characterized in that the substrate is a semiconductor wafer of over-insulating Semiconductor type, comprising at least one of the following materials: Si, Ge, SiC, CdTe, CdZnTe, GaAs, B ₄ C. 4. Microdosimeter according to any of claims 1 and 2 characterized in that the substrate is a standard wafer. 5. Microdosimeter according to any of claims 1 to 4 wherein the reading of the electrical signal is of the continuous and / or independent type for each of the cells. 6. Microdosimeter according to any of claims 1 to 5 comprising layers of various equivalent biological tissue materials. 7. Microdosimeter according to any of claims 1 to 6 comprising a signal processing system that obtains radiobiological quantities by means of an algorithm. 8. Microdosimeter according to claim 7 wherein the processing system correlates the variable measured by the microdosimeter, the energy deposited by ionizing radiation, ε, with radiobiological quantities such as that of the estimated dose equivalent H by an algorithm comprising geometric and tissue corrections equivalent. 9. Microdosimeter according to claim 8, wherein the algorithm comprises the following steps: a) measurement of the pulse height or load collected in each cell of the microdosimeter matrix as a spectrum or pulse height distribution in each pixel of the semiconductor that acts as substrate of the matrix that forms said microdosimeter, b) assignment of said pulse height of the corresponding energy deposited (ε) by the radiation in each cell, c) calculation of /, "average string length", defined as A- ^V  s being V the irradiated microscopic volume-sensitive and S the area of such volume,  _ ε d) calculation of linear energy, and, defined as y— ^~ j ^~ e) calculation of its linear energy distribution, f (y), f) calculation of the total absorbed dose distribution in the semiconductor as

being and the average frequency of the linear energy, calculated according to:

g) correction of tissue equivalence via the scale factor using the ratio of braking powers of the tissue and the semiconductor:  ^ Tissue  «Sil icón i h) calculation of the estimated dose equivalent convolving the data from the previous section with the quality factor, Q (LET) or Q (y) ■ dy

i) estimate of the relative radiobiological efficacy RBE, according to:

10. Method of manufacturing the microdosimeter defined in any one of claims 1 to 9 comprising the following steps: preparation of the semiconductor substrate by microelectronic or MEMS processes,  cylindrical-3D engraving on the semiconductor,  creation of metallizations on the P + and N + motifs to create the electrodes for reading the electric charge. 11. Method according to claim 10 wherein in step a) the central component type P + or N + is created by diffusion or implantation. 12. The method according to claim 11 wherein step b) comprises a cylindrical-3D ring-shaped engraving centered on the pixel of the cell, and is filled with doped polysilicon or undoped polysilicon which is subsequently doped with dopant type N + or P + respectively. 13. Manufacturing method according to claim 10 wherein in step a) a P + or N + type component is diffused or implanted or a cylindrical-3D-ring engraving is made filled with already doped polysilicon or undoped polysilicon and subsequently it is doped P + or N + type. 14. A method according to claim 13 wherein in step b) it comprises the formation of a cylindrical-3D column in the center of the cell, whose internal walls are filled with already doped polysilicon or with undoped polysilicon and later this is doped type P + or N + 15. A method according to claims 10 to 14 wherein the cylindrical etching process comprises a reactive beam process by ion beam or a deep etching process by reactive ions. 16. A method according to claims 10 to 15 wherein the cylindrical etching process comprises a process of thinning the substrate by chemical etching or reactive ion processes. 17. Use of the microdosimeter defined in claims 1 to 9 for radiation detection in the field of medical application. 18. Use according to claim 17 in hadrontherapy. 19. Use according to any of claims 17 to 18 for carrying out measurements at the surface level of the skin and / or at a biological interface, and / or at a certain depth such as under the skin or irradiated tissue. 20. Use according to any of claims 17 to 19 of the microdosimeter coupled to a tissue equivalent material that simulates specific biological tissue including water, bone and / or muscle. twenty-one . Use according to any of claims 17 to 20 in other fields derived from other radiation environments, such as secondary neutron generated in radio / hadronterapia. 22. Use according to any of claims 17 to 21 in mixed radiation fields, including radiation protection in personal use and area monitoring, for portable systems, and in the field of aerospace application.