IT202100000605A1 - RADIATION-RESISTANT WEARABLE SILICON DOSIMETER BASED ON FLOATING GATE SENSOR TECHNOLOGY - Google Patents

Description

RADIATION RESISTANT WEARABLE SILICON DOSIMETER BASED ON FLOATING GATE SENSOR TECHNOLOGY

TECHNICAL FIELD

The present invention ? relating to a device for determining individual exposure to ionizing radiation. In particular, the present invention relates to a radiation-resistant and reusable wearable dosimeter, comprising a radiation sensor based on the principle of the discharge of a Floating Gate (FG) MOS (Metal Oxide Semiconductor) transistor, coupled to a processing, both made in radiation resistant technology (Radiation Hardened By Design, RHBD), and integrated monolithically in a semiconductor device made in CMOS technology (Complementary Metal Oxide Semiconductor), said dosimeter being usable in various applications, such as radiotherapy .

BACKGROUND

Dosimeters are used for medical purposes to protect patients subjected to ionizing radiation. In fact, the exposure of the biological tissues of a part of the human body to a high dose of ionizing radiation can determine forms of poisoning, potentially lethal. Small wearable dosimeters are therefore used, which allow you to measure the dose to which ? subjects a patient during radiation exposure in therapeutic treatments and medical diagnostics.

However, despite the equipment used in therapies are increasingly? precise and more technologically advanced, think of the latest generation radiotherapy equipment in use today for oncological treatments, the precision of the dose imparted compared to the expected one remains of the order of ? 5%. Bearing in mind that approximately 7.5 million radiotherapies are performed each year worldwide, there could potentially be a large number of cases in which a discrepancy in the dose of radiation administered compared to that expected could lead to a very dangerous overdose in patients treated with radiation for therapeutic.

The problem lies in the fact that inaccurate, complicated or even obsolete technologies are often used in the evaluation of absorbed doses. For example, a technique used for dosimetry in radiotherapy ? film dosimetry. It uses detectors made up of highly sensitive planar films, with which the radiation interacts. This set includes, for example, radiographic films, which are made up of an emulsion of silver salts and gelatin on a rigid support, and require post-irradiation development (chemical process). Therefore, different process steps can affect the accuracy and reproducibility of the process. of the measure. Or radiochromic detectors, which are transparent films consisting of a single or double thickness of radiation-sensitive material. They respond to radiation by turning blue following a polymerization process of monomers of organic single crystals.

However, film-dosimetry ? a complex technique and a comparison between different results pu? only be carried out under the same measurement conditions.

Greater reliability? and reproducibility of the measure is obtained using the technology of the MOSFET transistors (Metal Oxide Semiconductor Field Effect Transistor) integrated in the ionizing radiation dosimeters, as a result of their capacitance? to track the rays-? and X-rays. Furthermore, their small size makes it possible to measure the dose during applications of radiation for medical or diagnostic treatments, even in-vivo. MOSFET transistor dosimeters have been used for many years in space applications, especially to measure crew exposure during space activities. extra-vehicular.

The current technologies for the realization of absorbed radiation dosimeters are mainly based on the use of sensors based on the technology of the MOSFET transistors with Floating Gate (FG). This device bases its operating principle on the discharge of its threshold voltage which occurs when it is exposed to radiation. Once used for measurement, the sensor can? be recharged.

For example in the article ?A very-low-cost dosimeter based on the off-the-shelf CD4007 MOSFET array for in vivo radiotherapy applications? (O.F.Siebel et al., 1350-4487/? 2015 Elsevier Ltd.) a case of the use of a low-cost MOSFET dosimeter suitable for in vivo radiotherapy applications is reported. Furthermore, the fundamental aspects of the basic constituent elements of a MOSFET dosimeter, namely the radiation sensor, the readout circuit and the temperature dependence are discussed.

The radiation sensor? therefore one of the constituent elements of a MOSFET dosimeter. The conventional floating gate radiation sensor, described in US patent 4,788,581, bases its operation on the creation, during exposure, of electron-hole pairs (e-/h<+>) in the oxide film, which charge the floating gate structure, that is? an unconnected structure, isolated from the rest of the device. This method requires the application of a voltage while the detection occurs, which can? be problematic for applications requiring very low power operations. Furthermore, the fact of using a relatively thin oxide to convert the incident radiation into e-/h<+ >pairs can? be a limit to sensitivity? of the device.

For example, US patent 8,519,345 B2, published in 2013, claims a MOSFET sensor made in CMOS technology, based on the discharge of a field oxide capacitor, coupled to an injector transistor and a readout transistor. However, if on the one hand the sensor solves the sensitivity problems, using a thicker Field Oxide? high to convert the radiation into electron hole pairs, on the other hand it does not use the Control Gate (CG), thus resulting in a poor controllability? of the sensor during its operation.

In the patent application US2015/0162369 published in June 2015, a radiation sensor based on the Floating Gate (FG) technology is described, which solves the above problems, having a high sensitivity? to the ionizing radiation and an?excellent capacity? of control in the programming and reading phases. The sensor includes a Floating Gate (FG) structure having a large region containing both a Tunneling Capacitor (TC) and a Control Capacitor (CC), the capacitance of which is? ? up to 100 times that of TC, to provide excellent programming control and facilitate programming of the FG structure using ?Fowler-Nordheim tunneling? techniques. In particular, the CC is constituted by a layer of polycrystalline silicon arranged above a thick insulation region (such as, for example, a shallow trench insulation structure or Shallow Trench Isolation, STI), about 50 times more of the oxide layer of the Tunnel Effect Capacitor, and by a Control Gate (CG) made by means of an isolated region (P-Well) of P-type doped silicon (holes), arranged below the thick dielectric structure. Thick dielectric structure promotes high sensitivity to ionizing radiation providing a large sensitive volume which increases the number of electron-hole pairs (e-/h<+>) producing the floating gate discharge during exposure. Furthermore, this structure? realizable through a standard CMOS production process and can be integrated monolithically in a device together with an external reading circuit and an analog-to-digital converter. The monolithic approach appears convenient both in terms of performance and construction costs.

A radiation sensor of this type, therefore, ? certainly an excellent candidate to realize the design of a reusable, compact and economical wearable dosimeter, as it works at low voltage and does not require power during exposure (passive).

The signal reading and processing circuit? another building block of a MOSFET dosimeter. To ensure reusability? of the device, the reading circuitry must be designed in such a way as to be resistant to radiation (?RAD-HARD?) and to avoid the degradation of the internal circuitry due to the effect of the absorbed radiation, which could afflict the measurement of the dose itself. The floating gate radiation sensor can? be used more than once due to its characteristic of being rechargeable prior to exposure causing it to discharge, therefore the effects of circuitry degradation due to exposure must be taken into account. In medical applications, for example, you can? expect a total dose to exceed 100 Gy after a certain number of irradiation cycles. Besides the sensor itself, all electronics should be tolerant of radiation up to the maximum absorbed dose.

In conclusion, a dosimeter that has the following characteristics:

? is realized by means of a floating gate radiation sensor,

? has small dimensions, allowing to measure the dose during radiation applications,

? is able to measure the radiation dose quickly and accurately, ? both rechargeable and, therefore, reusable pi? once upon a time,

? functions as a passive dosimeter, and as such is capable of measuring integrated dose values over a long period of time, does not require power during radiation exposure and functions at low power,

? is made in a standard technology such as CMOS,

? is implementable in a single chip together with the reading and processing circuitry

? and whose components can be manufactured in a radiation-resistant manner (RAD-HARD),

? therefore desirable.

SUMMARY

The present invention refers to a reusable, monolithic and radiation resistant (?RAD-HARD?) wearable dosimeter, made in CMOS technology, including a floating gate passive ionizing radiation sensor integrated monolithically in a single chip together with a circuitry readout and a processing circuit for converting the output current into voltage and the analog output signal into digital.

According to an embodiment of the invention the wearable dosimeter consists of a totally integrated solution which can be reusable.

According to an embodiment of the invention the wearable dosimeter comprises a single sensor or an array of sensors manufactured on a semiconductor substrate, each sensor comprising a floating gate (FG) structure isolated from the rest of the device, said FG structure comprising a tunnel effect capacitor (TC), or injector, formed with a thin Gate oxide, to introduce the charge inside the structure FG, a Control capacitor formed with a dielectric about 30 times larger? often of the Gate oxide and which shares with the TC the floating Gate structure and a reading circuitry including PMOS and NMOS transistors used to make the current flow; said dosimeter further comprising a processing circuit comprising a programmable voltage-to-current converter (I/V) with resistive detection and an analog-to-digital converter (ADC) interfaced with said voltage-current converter (I/V).

According to an embodiment of the present invention the floating gate structure (FG), realized through a layer of polycrystalline silicon, is shared between the control capacitor (CG), the tunnel effect capacitor (TG) and the NMOS and PMOS transistors belonging to the reading circuitry.

According to an embodiment of the invention, the ADC converter belongs to the ADC typology (from 5 to 10-bit) based on the resistive string flash architecture, and is? manufactured using Radiation-Hardened-By-Design (RHBD) techniques, such as ?Enhanced Guard Ring? (EGR), ring transistors or ?Edgeless Transistor? (ELT), both in terms of architecture and layout, to ensure that the absorbed dose does not degrade the circuitry.

According to an embodiment of the present invention, the sensor is charged by injection of the charge into the floating gate (FG) structure according to the ?Fowler-Nordheim? tunnel effect mechanism? applying high positive voltage (for example 8V) to the Gate electrode of the Control Capacitor (CG) and negative (for example - 4V) to the Gate electrode of the Tunnel Capacitor (TG); once the maximum amount? of charge ? been stored in the FG (condition in which the current flowing in the circuit is zero), the sensor is exposed to the radiation; the energy released by the incident radiation allows the trapped charges to find an escape route and the FG is potentially progressively discharged until it is completely empty of charge carriers (condition of maximum current flowing in the circuit); these two extreme conditions correspond to a transition from a situation of zero absorbed dose to a situation of maximum detectable and measurable absorbed dose; the sensor is then recharged again to reinitialize the sensor under the same measurement conditions prior to irradiation.

According to an embodiment of the present invention, the sensor is programmed by applying high voltage (for example 8V) to the Gate electrode of the Control Capacitor (CG) and negative (for example - 4V) to the Gate electrode of the Effect Capacitor Tunnels (TG); the threshold voltage variation is checked by applying low voltages; when the threshold voltage ? elevated the dosimeter ? ready to be irradiated; as a result of the absorbed radiation, the threshold voltage is progressively lowered with respect to its initial value and the variation of the threshold voltage determines the increase of the current; the current variation connected to the threshold voltage variation is converted into a voltage value by a programmable voltage-current converter (I/V) and transformed into a digital value by an analog-digital converter (ADC), so that it can be read from any digital system.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1A shows a floating gate MOS transistor (1) according to the prior art. Figure 1B shows a radiation sensor (100) according to the prior art comprising an injection region (10), a sensitive region (20) and a processing region (30) and having charges trapped in the sensitive region.

Figure 2 shows a radiation sensor (100) according to the prior art during exposure to radiation (40) which determines the emptying of the charges (60) from the sensitive area (20), and a residue of charges remaining in the area sensitive (50). Figure 3 shows a radiation sensor (100) according to the prior art after exposure to radiation (40), containing remaining charges (70) in the sensitive area.

Figure 4 shows a radiation sensor (100) according to the prior art before exposure to radiation (40), in the condition of maximum filling of the charges (80) and minimum current.

Figure 5 shows a radiation sensor (100) according to the prior art after exposure to radiation (40), in the condition of total emptying of the charges (90) and maximum current.

Figure 6 shows the graphic representation (200) of the trend of the drain current Idr as a function of the threshold voltage Vt of a radiation sensor.

Figure 7 shows a block diagram of a wearable dosimeter (300) according to an embodiment of the present invention, comprising a radiation sensor (C-Sensor), (310), a voltage-current converter (I-V converter), ( 320) and an analog to digital converter (ADC), (330) made in radiation resistant technique.

Fig. 8 shows the circuit diagram (320B) of a voltage-to-current converter (320) according to an embodiment of the present invention.

Figure 8A shows the physical layout (350) of a pair of ring transistors of the Edge Less Transistor (ELT) type used in the architecture of the dosimeter.

Figure 8B shows the physical layout (400) of a region of the I-V converter (320) comprising a plurality of of switches, made using N-type (401) and P-type (403) ring transistors (ELT) separated by a double crown of guard rings (402 and 404).

Figure 8C shows a cross section (500) of a region of the I-V converter (320) comprising P+ bias areas containing P transistors (XP) and N+ bias areas containing N transistors (XN), separated by guard rings (510A and 510B) with substrate (511) and Nwell (512) bias.

Figure 9 shows the circuit diagram (600) of an analog to digital converter ADC (330) according to an embodiment of the present invention.

Figure 10 shows the trend of the characteristic curve (650) of the Floating Gate MOSFET sensor before (before) and after (after) the irradiation.

Figure 11 shows the graph (660) of the trend of the threshold voltage variation (Voltage Shift) as a function of the absorbed radiation (rad) for a series of sensors compared (#1,3,7 vs #2,5, 10).

DETAILED DESCRIPTION

The idea of using MOSFET transistors for integrated dosimeters? been developed since the 70s and ? been applied in many fields, such as, for example, space, nuclear power plants and the military, to then be applied also in that of medical therapy and diagnostics. As far as space applications are concerned, MOSFET dosimeters have been used for many years to verify the exposure to cosmic rays of crews in extra-vehicular expeditions. Traditional MOSFET dosimeters are based on FGMOS technology, or ?Floating Gate? (Fig.1A). A traditional FGMOS (1) consists of a standard MOS to which ? the Floating Gate (FG) has been added, i.e. a layer of Polysilicon (6) electrically isolated by two layers of Silicon Oxide (2), between the semiconductor substrate, e.g. Silicon (7), and the Control Gate (CG) (8). The insulating layer of Silicon Oxide (2) which separates the FG from the substrate must be thin enough to allow the injection of charges into the substrate, while the layer which separates it from the Control Gate must be thick enough not to allow the leakage charge from the device. There are numerous variations of this scheme, some of which have the addition of more? gate terminals for the transistor in order to be able to store more? of a bit of information.

To charge the FG (6), a positive voltage must be applied to the CG (8) to pull the electrons from the channel (5) to the Floating Gate (6). Clearly first the electron channel (5) must have been formed by applying a potential difference between the source (3) and drain (4) electrodes.

An embodiment of a Floating Gate MOSFET radiation sensor is presented in patent application US2015/0162369, used hereinafter as reference, in which the radiation sensor based on a Floating Gate MOSFET non-volatile memory cell comprises a ?Tunnel Gate?, separated from the substrate by a thin insulating layer (silicon oxide) and a ?Control Gate?, in which the insulating region ? about 30 times more thick.

According to an embodiment of the present invention, the monolithic wearable dosimeter and made in radiation-resistant technique comprises a radiation sensor (100) of the Floating Gate MOSFET type described in patent application US2015/0162369, to which express reference is made in the this description. The radiation sensor (100), schematically shown in Figure 1B, comprises a Floating Gate (FG) structure, made of polysilicon, (12), which ? shared between 3 regions: a charge injection region (10), a radiation sensitive region (20) and a signal processing region (30). In the injection region (10), the Floating Gate structure (12) ? separated from the silicon substrate by a layer of insulator, silicon oxide, with a thickness between 110 and 150 nm, which ? the Gate Oxide (11). The capacitor made up of the polysilicon of the Gate electrode (12), the Gate Oxide (11) and the P-well region created in the underlying Silicon substrate, forms the ?Tunnel Capacitor (10B)?, or effect capacitor Tunnel, through which the charges are injected into the sensitive region (20) of the sensor. In the sensitive region (20) the ?Floating Gate? ? separated from the underlying Silicon region (P-well) by a thick layer of insulator, for example a surface isolation trench Silicon Oxide (14), or STI ?Shallow Trench Isolation?, often 30 times the Gate Oxide ( 11), i.e. of a thickness of about 3500 nm, in which the charges are stored and are unable to escape until they receive sufficient energy to overcome the forces that hold them inside the FG layer. The capacitor made up of the Gate electrode (12), the STI Oxide (14) and the Pwell region of the Silicon substrate ? the Control Capacitor or ?Control Capacitor? (20B), whose extended volume allows an increase in stored charge and therefore in sensitivity. Finally, the signal processing region (30) ? consisting of a reading circuitry comprising a series of P-type (PMOS), i.e. P-channel (holes), and N-type (NMOS) or N-channel (electrons) MOSFET transistors (13). The voltage applied to the gate electrode of said MOSFET transistors (13) in the reading circuitry is controlled by the gate electrode (15) of the control capacitor (20B).

As previously mentioned, the physical principle on which the sensor is based? the discharge of the floating gate region due to the absorption of a dose of ionizing radiation. Initially the sensor is programmed by applying a positive voltage (about 8V) to the Gate electrode (15) of the Control Capacitor and a negative voltage (-4V) to the ?Tunnel Gate? (16). Thanks to the reduced thickness of the Gate Oxide, this potential difference is sufficient to trigger a ?tunnel effect?, through which the charges penetrate the structure to Floating Gate (FG), which ? isolated from the rest of the device by silicon oxide. Once stored in the FG structure, the charges have no chance? to escape if not through the? application of a? high voltage in mode? reverse or through other mechanisms capable of transferring enough energy to the charges to make them escape: among these mechanisms c?? the one that exploits the energy of ionizing radiation.

In fact, the energy released by the incident radiation allows the trapped charges to climb over the potential barrier present at the interface between the silicon substrate and the silicon oxide and make them escape to reach the transistor channel. For example, in Fig 1B the radiation sensor is shown having charges trapped at the sensitive region (20). In Fig.2 you can? observe how due to the effect of the incident radiation (40), some charges (60) receive sufficient energy to escape the potential barrier which keeps them trapped in the FG and a progressive process of discharge of the FG is determined. Once the irradiation is finished, (Fig.3), there remains a residual charge (70) stored in the FG structure (12) corresponding to a certain quantity? of absorbed radiation, which charge is calculated by the difference between the quantity? of charge stored before and after exposure. Once the FG is completely emptied of the charges, the sensor is in the maximum current condition that the MOSFET allows to deliver (Fig.5), i.e. drain current Id = Imax; conversely, in the initial conditions the current delivered was practically zero, ie Id = 0 (Fig.4). The passage from a condition of null current supplied to the condition of maximum current supplied determines the transition from a situation of zero absorbed dose to the situation of maximum absorbed dose (measurable). Recharging the sensor? It is possible to repeat another measurement under the same measurement conditions.

The variation of the current? directly related to the variation of the threshold voltage (Vt) of the MOSFET, or minimum voltage applicable to the gate of the MOSFET to make the current flow through the channel. In fact, due to the effect of the incident radiation, the net charge stored in the FG is ionized, with the creation of pairs of holes<+>/electrons-. By applying a positive voltage to the Control Gate, the charges are separated, i.e. the electrons are attracted towards the Gate electrode and carried away by the positive potential difference, the holes are directed towards the Silicon substrate. The accumulation of holes, or positive charges, in the substrate causes an increase in the threshold voltage (Vt) required for current to flow through the electron channel of the MOSFET. The difference in threshold voltage with and without irradiation ? related to the imparted dose with a high degree of accuracy. For example, the value of the threshold voltage ? obtainable from the value of the Drain current, Id, which flows through the N-channel MOSFET (NMOS), while the P-channel MOSFET (PMOS) is connected to ground (turned off).

For example, in figure 6 ? the graphic representation of the trend of the drain current (Id) as a function of the threshold voltage (Vt) is shown.

The simple variation of the current, that ? directly connected to the variation of the threshold voltage, and coming from the sensor, not ? clearly ready to be digitally processed, but requires a conversion first from current value to voltage value and then from analog value to digital value, to obtain an output bit stream that can be used by any digital system.

According to an embodiment of the present invention the output current of the sensor is processed by a programmable voltage-current converter (I/V) and by an analog-digital converter (ADC).

For example, in Fig.7 ? shown is a block diagram of the monolithic wearable dosimeter (300) made according to an embodiment of the present invention, comprising a radiation sensor (310), an I/V converter (320) and an ADC converter (330). According to this scheme, the current Id coming from the sensor is processed by the I/V converter and transformed into a voltage value, Vout, which is, in turn, sent to the analog-digital converter for transformation into a string of bits (or sequence of 0, 1).

According to a preferred embodiment of the present invention, with the exception of the sensor itself, all the electronic circuitry which constitutes the conversion chain of the wearable dosimeter ? made using the ?Radiation Hardness By Design? (RHBD) in such a way as to be tolerant to incident radiation up to the maximum total absorbed dose (TID). In particular, the RHBD technique used ? focused on mitigating the effects of ?Total Ionization Dose (TID)? and countermeasures to avoid damage from single ?latch-up? (SEL) circuits. Indeed, how? known, one of the major risks of CMOS technology? the trigger of the ?latch-up?, caused by the presence of a parasitic circuit inside the basic architecture of the technology. The effects more relevant in CMOS circuits due to the total absorbed dose can be both the degradation of the threshold voltage and the triggering of parasitic leakage currents between Source and Drain in the MOSFET transistors.

For example, in the present invention it is resorted to the transistors ?Enclosed Layout? (ELT), which help maintain a low level of leakage current in N-MOS transistors even after irradiation. ELT transistors are also known as edgeless, or edgeless, transistors because their design avoids including the edges of the deep trench insulation structure, or STI, between the source and drain electrodes of the MOS transistor. Among the versions used of ELT transistors are there those with ?enclosed-gate? (Fig.8A), in which the Gate electrode (351) of the transistor has the shape of a ring and completely surrounds the Source (or Drain) contact so that no leakage channels can form between the Source (352) and the Drains (353). The passive elements were also chosen according to the radiation tolerance requirement.

According to an embodiment of the present invention the I/V converter (Fig. 8) is realized through the implementation of two cascade amplifiers, the first of which ? a ?mirrored cascode? (315) or two-stage amplifier and the second a level shifter or ?level shifter? (316). Compared to a single amplifier stage, this combination provides greater input-output isolation, higher output impedance, and greater bandwidth. Also, with its high output impedance, a ?cascode? can? function as a high gain amplifier. The ?cascode? ? consists of two transistors, one of which acts as a common source and the other as a common gate. The level shifter (or ?level shifter?) in digital electronics, ? a circuit used to translate signals from one logic level, e.g. voltage domain to another, e.g. current domain.

The I/V Converter (320B) shown in Fig.8 benefits from various expedients all aimed at mitigating the effects of ionizing radiation; for example polycrystalline silicon resistors are used in the feedback chain (321B), which are not only stable when irradiated (active reading), but which do not degrade as a result of the total absorbed radiation dose. Said resistors are furthermore connected to switches S1,?, SN, (322B) constructed according to RHBD rules; these consist of an N-channel transistor and a P-channel transistor both built with ring geometry (ELT) to obviate the degradation along the edges of the channel following the generation of electron-hole pairs during irradiation.

According to an embodiment, said switches, which are used in the compensation chain, are provided with a double polarization ring, as shown in Fig.8B wherein ? shown is a top view (400) of the double crown of guard rings, with respectively substrate and Nwell biases, which separates the area of the P-type transistors (ELT) (403) from that of the N (401). The substrate bias guard ring (402) surrounds the zone containing the N-type transistors (401) and the Nwell bias guard ring (404) surrounds the zone comprising the P-type transistors (403).

In Fig.8C the areas containing the P transistors (XP) and the N transistors (XN) separated by guard rings (510A and 510B) with substrate (511) and Nwell (512) bias are shown in cross section. In this way the N-channel transistors (XN) ?see? their P-channel counterparts (XP) through two levels of guard rings: the one belonging to their bias (substrate bias) and the one belonging to the bias of the P-channel transistors (NWell bias). Since the substrate guard ring (510A) is? connected to mass (Gnd) and the? guard ring of Nwell (510B) ? connected to the supply voltage (VDD), a strong field inversion zone is generated, which prevents the passage of any parasitic currents, such as those deriving from the degradation of surface trench oxides (STI), which normally separate the P-type zones from those of type N. In normal conditions, in fact (without guard rings), ? It is known that the accumulation of holes is generated in correspondence with the trench oxides (STI), which constitutes a current of parasitic holes which puts the N zones in communication with the P zones.

This type of guard ring solution ensures stability towards any type of degradation in the oxides, a fundamental condition in the context of the conversion of an analog quantity such as that coming from the threshold voltage of the sensor. Similarly, in the design of operational amplifiers (315, 316), the RHBD techniques adopted have been directed towards a mitigation of the degradation effects in thick oxides. Also in this case, said operational amplifiers are designed in such a way that the N-channel transistors are physically separated from the area in which the P-channel transistors are positioned and are surrounded by guard rings.

According to an embodiment of the present invention the analog-to-digital converter ? realized in form of ?flash ADC?. The flash architecture of the ADC ? was chosen precisely for its proven tolerance to radiation up to a total absorbed dose of 3 kGy and its insensitivity? at single ?latch-up? events of the circuit. According to a preferred embodiment, the ADC converter ? built using a? Flash architecture, which ? represented according to the circuit diagram (600) in Fig.9. According to this embodiment, called Flash architecture ? based on a resistive string, made up of 2<N >resistors (631) (in the case of N bit ADCs) and 2<N >-1 comparators (632). The unknown voltage Vin is compared to a reference voltage VR, which is to be compared with the one to be converted. Before entering the comparators, the reference voltage is divided through 2<N> resistors providing each comparator with an input value more? smaller than the next one by a value that can be associated with the least significant bit or LSB. The digital word made up of the comparator output bits represents the digital conversion of the analog voltage present at the converter input. At the output of each comparator ? there is a regenerative latch that stores the result. The latch has a positive feedback to ensure that the data stored is a "1" or a "0" to avoid metastability phenomena.

The ADC converter is also made using RHBD technology, providing two levels of protection against ionizing radiation: a first level which concerns the architecture and a second which concerns the physical design. The choice of a Flash architecture based on a resistive string is based on the principle of the intrinsic resistance that the resistive layers built in polycrystalline silicon show against ionizing radiation. In this context, where radiation is moderate with respect to a purely spatial environment, isn't it? both the malfunction or the possible breakage of the device that must be remedied, as its stability? within the conversion chain. The Flash architecture? simple, extremely robust and is based on building blocks (the comparators) that are easily implemented with RHBD techniques compared to other analog structures, certainly more? refined, but also prone to radiation degradation.

The comparator (632), key element of the ADC, ? made up of a series of inverters, each one fed back through a ?switch? and by MIM (Metal-Insulator-Metal) capacitors that sample the input reference voltage, partitioned by the resistive matrix. The fact that the comparison is made through inverters in turn made in RHBD technology (using ELT transistors, equipped with guard rings), makes the system particularly robust, and the use of MIM capacitors, also insensitive to the total dose of radiation, further contributes to making the structure less sensitive to radiation. The architectural aspect is also supported by an accurate physical layout, in which in addition to the traditional ELT transistors, a massive use of guard rings avoids any possible polarization coming from the entrapment of holes in the insulation oxides; in particular in the ?switches? sampling points which must transfer the analog quantity (the partitioned input voltage) to the comparator.

Analogous approach with guard rings ? adopted for MIM capacitors. A crown of insulation is placed around each capacitor to avoid any possible polarization effect coming both from neighboring capacitors and from other active elements, such as ?switches?.

According to an embodiment of the present invention the wearable dosimeter, including the radiation sensor, the I/V converter and the flash ADC converter ? feasible in standard CMOS technology on an electronic device in monolithic form, for example, can you? use 180nm CMOS technology to make a chip whose main area ? less than 0.25 mm<2 >and in which the sensor occupies an area no longer? large than 20 x 20 microns.

Operating and performance parameters.

To charge the Floating Gate sensor, the charge must be conducted in such a way as not to damage the physical device and for this reason it is carried out by applying a train of high voltage pulses and verifying the current actually supplied at the end of each pulse. For example, a range of voltages applied to the Drain electrode ranging from 0.5V to 1V is used, keeping the voltage applied to the Gate electrode fixed at 5V. The threshold voltage of the sensor at the end of the charging procedure measures, for example, Vt = 3.5 V, corresponding to a drain current of about 40?A. Then exposing the sensor to different doses of ionizing radiation, ? It is possible to carry out a measurement of the discharge of the threshold voltage (Vt). Before irradiating the sensor, we proceed to measure the linearity? of the I/V converter, which must be < 3%. For example, in a range of 0.4V ? 3,9V the transfer function, Vin Vs Vout, of the I/V converter shows a linearity? < 2%. After a series of charge and discharge cycles, the transfer characteristic shows no substantial changes.

In the characterization phase, the dosimeter ? been irradiated with an irradiated dose rate of 1.44 Gy/min (Si), in a dose range of 3 to 10 Gy (Si). In Fig.10 ? the characteristic curve (650) of the sensor before (before) and after irradiation (after) is shown, which shows a decrease in the threshold voltage of 0.75 V after irradiation. After reading, the dosimeter ? been recharged and irradiated again to check the possibility? of reuse. Furthermore, for each dose, 3 dosimeters were irradiated to verify the repeatability. of the measure. Fig. 11 shows the graph (660) of the threshold voltage variation, or ?Voltage shift? as a function of the irradiated dose for 2 different series of three sensors. Can you? note how the ?voltage shift? show good linearity? in the range 0.25-0.7 V and a good repeatability? among the series of sensors subjected to irradiation, evaluated to be for each dose value within a range of deviation from the average value no longer? 6% large.

Furthermore, still in the characterization phase, the power consumption of the dosimeter ? been verified not to exceed the value of a 2 mW at a supply voltage of 5V. What about sensitivity? the dosimeter ? capable of detecting a radiation dose up to 10 Gy with a resolution of 0.3 Gy in a temperature range from 0 to 85 ?C.

In conclusion, the Floating Gate Dosimeter according to an embodiment of the present invention has the following characteristics:

? ? based on a floating gate memory programmed by tunnel effect (Fowler-Nordheim);

? during charging the threshold voltage is brought to high values;

? the irradiation that occurs in mode? passive (no power supply) with floating gate sensor causes sensor discharge;

? in mode? activates a conversion chain composed of an I-V converter interfaced with an ADC translates the variation of the threshold voltage into digital values;

? both the I-V converter and the ADC are made in RHBD technology, both in terms of architecture and physical design, through the implementation of polycrystalline silicon resistors, ring transistors (ELT) and insulation zones made with double ring chains of guard;

? the dosimeter is integrated in standard 180 nm CMOS technology;

? tests performed on pi? dosimeters loading and unloading pi? times the sensor show good linearity? of response (<2%) and an?excellent repeatability? (<6%). The advantages offered by the present invention are clear from what has been explained above. In particular, ? important to underline the fact that, since? the charge process of the Floating Gate must be repeated more? once and that in medical applications this fact pu? result in a total dose which, at the end of a series of irradiation cycles, can exceed the 100Gy, to ensure the reusability? of the dosimeter, both the I/V converter and the ADC converter have been designed in RHBD technique in order to be tolerant to radiation up to the maximum exposure dose. The I/V converter coupled to the ADC conceived in this modality? RHBD, shows stability? towards radiation and resistance to degradation effects. Also, the fact that the dosimeter works in mode? passive, i.e. in the absence of power supply, makes it particularly easy in laboratory operations for operators in the sector and its small size allows precise positioning in the points most exposed to radiation in the case of patients subjected to radiation for therapeutic purposes. In the end, ? it is clear that numerous modifications and variations can be made to the present invention, all falling within the scope of protection of the invention, as defined in the attached claims.

Claims (16)

1. Monolithic, reusable, radiation resistant, wearable dosimeter comprising a rechargeable radiation sensor (100, 310), based on the principle of the discharge of a Floating Gate capacitor due to the effect of ionizing radiation, said sensor comprising: - a semiconductor substrate (for example silicon); - a Floating Gate (FG) structure (12), isolated from the semiconductor substrate and shared between an injection region (10), a sensitive region (20) and a processing region (30), called injection region ( 10) comprising a Tunnel effect Capacitor (10B) for injecting the charges into the Floating Gate structure (12), said sensitive region (20) comprising a Control Capacitor (20B), said capacitor being subject to loss of charge due to the ionization, and said processing region (30) comprising a reading circuitry (30B) comprising MOSFET transistors (13) of the NMOS and PMOS type; characterized in that it further comprises a conversion chain which includes an I/V voltage-current converter (320) for transforming the sensor output current into a voltage signal and an ADC converter (330) for transforming the analog input signal into digital output signal, said I/V converter (320) and said ADC converter (330) comprising elements made in Radiation-Hardened-By-Design (RHBD) technology to ensure that the absorbed dose does not degrade the circuitry and to mitigate the effects of Total Ionization Dose (TID) and prevent Single Event Latch-up (SEL). 2. Wearable dosimeter according to claim 1 characterized in that said Floating Gate structure (12) being separated from the Silicon substrate by a layer of Silicon Oxide, or Gate Oxide (11), comprised between 110 nm and 150 nm in correspondence with said Tunnel Effect Capacitor (10B) and by a layer of shallow trench insulator (STI), (14), with a thickness of about 3500 nm, in correspondence with said Control capacitor (20B). 3. Wearable dosimeter according to any one of the preceding claims wherein said dosimeter being programmable before the exposure to be reused more? than once. 4. Wearable dosimeter according to any one of the preceding claims, said sensor (100, 310A) comprising external electrodes (15, 16) for programming the sensor before irradiation by applying a positive voltage to the electrode (15) connected to the Gate of the Control Capacitor (20B) and a negative voltage to the electrode (16) connected to the Gate of the Tunnel Capacitor (10B), the application of said voltages determining the injection of the charge by tunnel effect through the Gate Oxide ( 11), and its storage inside the Floating Gate structure (12) in correspondence with the Control Capacitor (20B). 5. Wearable dosimeter according to claim 4, said positive voltage applied to the electrode (15) connected to the Gate of the Control Capacitor (20B) being equal to 8V and said negative voltage applied to the electrode (16) connected to the Gate of the Tunnel Capacitor (10B) being equal to ? 4V. 6. Wearable dosimeter according to any of the preceding claims, said reading circuitry (30B) being configured to read the residual charge (70) stored in the Floating Gate structure (12) of said sensor after exposure to an amount? of radiation absorbed by said sensor, said residual charge being determined by the difference between the quantity? of charge stored before and after exposure. 7. Wearable dosimeter according to claim 6, said reading circuitry (30B) being configured to supply an output current value, or Drain current, of said NMOS type MOSFET transistors (13), said output current being determined from the quantity? of total radiation absorbed by said sensor during exposure. The wearable dosimeter according to claim 7, said conversion chain, comprising said I/V voltage-current converter (320) and said analog-to-digital converter ADC (330), being configured to process said output current from said reading circuitry (30B). The wearable dosimeter according to any one of the preceding claims, said I/V converter (320) comprising a combination of first and second cascaded amplifiers, said first amplifier being a ?mirrored cascode? (315) or two-stage amplifier, and said second amplifier being a level shifter? (316), said combination by converting the input current signal to the output voltage signal. 10. Wearable dosimeter according to any one of the preceding claims, said elements made in Radiation-Hardened-By-Design (RHBD) technology comprising: polycrystalline silicon resistors, MOS transistors of the ring geometry type (Enclosed Layout Transistor, ELT), double crowns of Nwell and substrate biased guard rings, called crowns insulating the P-type MOS transistors from the N-type ones. 11. Wearable dosimeter according to any one of the preceding claims, said analog-to-digital converter (ADC), (330), belonging to the flash ADC category (from 5 to 10-bit) and comprising a resistive string of 2<N>resistance in series and 2<N >-1 comparators and subsequent logic circuitry for providing a digital output value. 12. Wearable dosimeter according to any one of the preceding claims, wherein the voltage applied to the Gate electrode of said MOSFET transistors (13) of said reading circuitry (30B) being controlled by means of the Gate electrode (15) of the Capacitor of Control (20B). 13. Wearable dosimeter according to any one of the preceding claims characterized in that it can be produced in standard CMOS technology in a totally integrated solution, or single chip, said chip having a main area lower than 0.25 mm<2>, inside said main area said radiation sensor occupying a surface area not exceeding 20 x 20 microns. 14. Wearable dosimeter according to any one of the preceding claims characterized in that it provides a dose measurement with a repeatability? included in an interval of deviation from the average value no longer? 6% large. 15. Wearable dosimeter according to any one of the preceding claims characterized in that it has a power consumption lower than 2 mW at a supply voltage of 5V. 16. Wearable dosimeter according to any one of the preceding claims characterized in that it detects a radiation dose up to 10 Gy with a resolution of 0.3 Gy in a temperature range between 0 and 85°C.