The present invention concerns a solid-state radiation sensor for the detection of radiations, in particular of particles.

Likewise, the method for the detection of radiations, in particular of particles, by means of the aforesaid solid-state radiation sensor is also part of the invention.

It is well known that in high-energy physics it is fundamental to reconstruct the trajectory of the charged particles produced as a result of an interaction with a resolution that can reach 10 pm in the most sophisticated experiments.

To achieve this precision, two fundamental requirements must be met: the segmentation of the sensor must be such as to achieve the required precision, and the thickness of the detector (hereinafter referred to as “material budget”) must be low enough to prevent multiple scattering from excessively worsening the trajectory estimate.

These requirements are partially met by the detectors in silicon or with semiconductor in general, which can be made with a pitch of a few tens of pm and with thicknesses of a few hundred pm. Silicon has a low ionisation energy, 3.6 eV, which makes it possible to obtain a detectable signal despite the fact that the thickness of the crossed material is relatively modest. These characteristics make silicon sensors very performing in tracking charged particles to reconstruct the decay vertices of short-lived particles, which decay very close (between 30 and 300 pm) to the beam interaction point. In addition, the segmentation of the sensor permits its operation even with high multiplicity of particles present in the regions close to the interaction point.

It is also known that, at present, there are essentially two types of silicon pixel sensors: the so-called in jargon hybrid pixel sensors and the active monolithic pixel sensors.

The first type of pixel sensors are devices in which the detection element and the reading electronics are made separately and then interconnected by means of the bump-bonding technique so as to form a single module. The active zone of the sensor is formed by a high-resistivity silicon layer that is i generally a few hundred pm thick, which, through the application of a voltage, becomes a depleted region allowing the efficient collection of charge. When a charged particle passes through one of the pixels, it generates a number of charge carriers. If these are produced in the depletion region, they result in a current pulse above the background signal (noise), and thus produce a detectable signal. The depletion region is therefore the active zone of the detector.

The reading electronics comprises an analogue front-end circuitry directly connected to the sensitive node and configured to perform analogue/digital conversion of the signals, followed by a reading logic for transmitting the output data to the detector. This latter portion of logic is not distributed over all the pixels, but is normally located near the edge of one of the chips and manages the information of multiple pixels.

However, disadvantageously, this type of hybrid sensors has a first drawback due to the non-negligible minimum sizes of the pixels, mainly caused by the use of analogue electronics, as well as by physical limitations related to the bump-bonding connection, the sizes of which cannot be smaller than 2025 pm, so that the pixel pitch cannot be decreased further.

Furthermore, the high size of the pixels also results in the definition of equally high capacities which in turn, disadvantageously, increase the noise level.

In addition, further technical disadvantages of such a hybrid approach consist of high current consumptions and fairly sophisticated circuitry.

There is also the disadvantage of the “material budget”: the presence of two layers (one dedicated to reading and one to the detection) disadvantageously increases the probability of scattering of the particle due to interaction with the detector itself.

The second known type of monolithic type radiation sensors are devices in which the sensor and the reading electronics are both made on the same low resistivity silicon substrate using the CMOS standard industrial process. Having the sensor and the reading electronics on the same substrate helps reduce the amount of material with which to make the device, reducing multiple scattering and therefore improving resolution, and at the same time simplifies the assembly of the detector and makes it more robust and reliable, because the hybridization process is not necessary, and also less expensive. By polarising the substrate appropriately, the monolithic detectors can achieve complete silicon depletion, thereby increasing performance in terms of collection rate and radiation resilience.

However, even this second type of approach involves the use of a portion of analogue circuitry that has an important impact on the size of each pixel. Also in this case, there is a high circuit complexity with relative power consumption. The present invention intends to overcome all the drawbacks and limitations of the known art that have been listed.

In particular, a first object of the invention is to realize a sensor of radiations, in particular of particles, with the sensitive elements (pixels) having a much smaller size than the sensitive elements of the radiation sensors of the known type.

This therefore makes it possible to realize a radiation sensor with a higher spatial resolution than the radiation sensors of the prior art with the same sizes.

A further object of the invention is to realize a purely digital sensor. This, as a first effect, allows the reduction of complexity in terms of the number of transistors and controls.

Furthermore, a further object of the invention is to realize a radiation sensor capable of maximising, in perspective, thanks to the digital approach, the technological scalability of its constituent components.

Furthermore, an object of the invention is to realize a radiation sensor that outputs an amount of reduced information of a digital type, which therefore do not need conversion and which are quickly processable and storable, while maintaining a high spatial and temporal resolution.

Last but not least object of the invention is the realization of a radiation sensor with reduced consumptions.

The objects listed are achieved by the radiation sensor of the invention according to claim 1 , to which reference will be made.

Other characteristics of the invention are described in the dependent claims.

The method for the detection of radiations, in particular of particles, by means of the aforesaid radiation sensor of the invention, according to claim 11 , to which reference will be made, is also part of the invention.

The objects listed will be better highlighted below during the description of a preferred, but not exclusive, embodiment of the invention which is given hereinafter with reference to the attached drawings in which: - fig. 1 schematically depicts a radiation sensor according to a preferred embodiment of the invention;

- fig. 2 depicts the circuit diagram of a pixel belonging to the radiation sensor of fig. 1 ;

- fig. 3 depicts the topographic section of a pixel belonging to the radiation sensor of fig. 1 ;

- fig. 4 depicts a time graph of the operating cycle of the radiation sensor of the invention;

- fig. 5 depicts the characteristic graph of the value of the critical charge in relation to the value of the voltage AV obtained with the radiation sensor of the invention.

The solid-state radiation sensor of the invention, for the detection of radiations, in particular of particles, is schematically depicted in fig. 1 , where it is indicated as a whole with 1.

According to the preferred embodiment of the invention, such a solid-state radiation sensor 1 comprises a plurality of radiation-sensitive pixels 3, made on a semiconductor substrate 2, preferably a silicon substrate, arranged in a matrix so as to define a plurality of columns 31 along a first direction X and a plurality of rows 32 along a second direction Y.

It is not excluded, however, that such a radiation sensor 1 , as a minimum configuration, may comprise a single pixel 3.

With regard to each pixel 3, as schematically depicted in fig. 2 and in the topographic section of fig. 3, it comprises a first detection transistor 4 of the n-channel MOSFET type, having the source terminal 41 connected to a lower voltage reference Vlow and the drain terminal 42 connected to a first supply line 5 of voltage Vdd by interposition of a first electronic switch 6, controlled by means of a first reset line 7.

Furthermore, said pixel 3 comprises a second detection transistor 8 of the p-channel MOSFET type, having the source terminal 81 connected to a higher voltage reference Vhigh, the drain terminal 82 connected to a second supply line 9 of voltage Vgnd, by interposition of a second electronic switch 10, controlled by a second reset line 11, and the gate terminal 83 connected to a first common node A with the drain terminal 42 of the first detection transistor 4.

Finally, the first detection transistor 4 has the gate terminal 43 connected to a second common node B with the dram terminal 82 of the second detection transistor 8.

In particular, as can be noted in the diagram of fig. 2, the first common node A is connected exclusively to the gate terminal 83 of the second detection transistor 8, the drain terminal 42 of the first detection transistor 4 and the first supply line 5 of voltage Vdd, by means of the aforesaid first electronic switch 6, while at the second common node B there are connected exclusively the gate terminal 43 of the first detection transistor 4, the drain terminal 82 of the second detection transistor 8 and the second supply line 9 of voltage Vgnd, by means of the aforesaid second electronic switch 10.

The technical advantage of this arrangement will be explained below during the description of the operating steps of the radiation sensor 1 of the invention. With regard to the voltage Vdd and the voltage Vgnd, the former is set equal to the higher voltage Vhigh with the addition of a voltage value AV, while the voltage Vgnd is set equal to the lower voltage Vlow minus the voltage value AV. According to the preferred embodiment of the invention, in particular, the value of the voltage AV is chosen greater than 0V. Also in this case, the advantages of such a deviation between the higher voltage Vhigh and the voltage Vdd and between the lower voltage Vlow and the voltage Vgnd will be indicated below during the description of the operating steps of the radiation sensor 1 of the invention.

Even more preferably, according to the preferred embodiment of the invention, the voltage value AV is between 50 mV and 300 mV, more specifically between 150 mV and 200 mV.

Functionally, it is important to point out that the value of AV determines the discrimination threshold of the event as well as, therefore, its sensitivity to detection.

It is not excluded, however, that according to an alternative embodiment this voltage value AV is imposed equal to 0V, so in this case the voltage Vdd corresponds to the higher voltage value Vhigh and the voltage Vgnd corresponds to the lower voltage value Vlow.

With regard to the higher voltage Vhigh and lower voltage Vlow values, they are respectively imposed, preferably but not necessarily, equal to a value between 1 ,8V and 0.8V, preferably 1 ,2V and a value around 0V.

Evidently, the value chosen for Vhigh depends on the technology in use. More dimensionally demanding technologies may require lower values for the higher voltage Vhigh than those indicated above for the specific embodiment example.

With regard to the first electronic switch 6 and the second electronic switch 10, according to the preferred embodiment, they are also two MOSFET transistors defined in the semiconductor substrate 2.

Even more specifically, the first electronic switch 6 is a p-channel MOSFET transistor connected with the drain terminal 61 to the first supply line 5 of voltage Vdd, with the source terminal 62 to the drain terminal 42 of the first detection transistor 4, then at the first common node A, and with the gate terminal 63 to the first reset line 7.

Similarly, preferably but not necessarily, the second electronic switch 10 is an n-channel MOSFET transistor connected with the drain terminal 101 to the second supply line 9 of voltage Vgnd, with the source terminal 102 connected to the drain terminal 82 of the second detection transistor 8, then at the second common node B, and with the gate terminal 103 to the second reset line 11. According to the preferred embodiment of the invention, as depicted in fig. 1 , the radiation sensor 1 of the invention comprises a plurality of first supply lines 5 of voltage Vdd, distinct from each other, and a plurality of second supply lines 9 of voltage Vgnd, distinct from each other. Even more preferably, the radiation sensor 1 of the invention comprises a number of first supply lines 5 of voltage Vdd equal to the number of columns 31 of the aforesaid pixel matrix 3 and, likewise, a number of second supply lines 9 of voltage Vgnd, always equal to the number of columns 31 of the aforesaid pixel matrix 3.

Each of the first supply lines 5 of voltage Vdd and each of the second supply lines 9 of voltage Vgnd is operatively connected exclusively with the pixels 3 belonging to one of the columns 31 of the pixel matrix 3. In this way, advantageously, it is possible to define for each column 31 of pixels 3 a specific value of the voltages Vdd and Vgnd, potentially distinct from the values of the voltages Vdd and Vgnd of the remaining columns 31 of pixels 3. In other words, in this way it is possible to vary and define the specific value of the voltage AV for each column 31 of pixels 3, independently of the remaining columns 31.

This advantageously makes it possible to independently determine for each pixel 3 present in the matrix the discrimination threshold of the event, as well as therefore its sensitivity to detection, thus compensating for any nonuniformities linked to threshold differences.

In an alternative embodiment, it is not excluded that the number of first supply lines 5 of voltage Vdd and the number of second supply lines 9 of voltage Vgnd may be equal to the number of rows 32 of the pixel matrix 3, rather than to the number of columns 31, and therefore it is not excluded that each of said first supply lines 5 of voltage Vdd and each of said second supply lines 9 of voltage Vgnd may be operatively connected exclusively to the pixels 3 belonging to a single row 32 of said pixel matrix 3.

Still, alternatively, a different embodiment of the radiation sensor 1 of the invention could provide a number of first supply lines 5 of voltage Vdd equal to the number of the pixels 3 of the matrix and, likewise, a number of second supply lines 9 of voltage Vgnd always equal to the number of the pixels 3, wherein each of the first supply lines 5 of voltage Vdd and each of the second supply lines 9 of voltage Vgnd is connected to one and only one of the aforesaid pixels 3.

Furthermore, a different embodiment of the radiation sensor 1 of the invention could provide a single supply line 5 of voltage Vdd and a single supply line 9 of voltage Vgnd, both connected to all the pixels 3 belonging to the aforesaid pixel matrix 3. Evidently, in this latter embodiment it will not be possible to distinguish the supply voltages Vdd and Vgnd imposed on the various pixels 3 of the radiation sensor 1 of the invention.

As regards, again, the radiation sensor 1 of the invention, it also comprises a logic unit 12 configured to perform in sequence and periodically:

- a pre-charging step 200, placing in operating mode the first electronic switch 6 and the second electronic switch 10, so as to impose on the first common node A the voltage Vdd and so as to impose on the second common node B the voltage Vgnd. Evidently, the state of operation of the first electronic switch 6 and of the second electronic switch 10 is imposed by the logic unit 12 respectively by acting on the two reset lines 7 and 11;

- a step of exposure 201 of the pixels 3 to radiations, placing in cut-off mode the first electronic switch 6 and the second electronic switch 10, so as to make the first common node A and the second common node B float. Also in this case, evidently, the cut-off state of the first electronic switch 6 and of the second electronic switch 10 is imposed by the logic unit 12 respectively by acting on the two reset lines 7 and 11;

- a step of reading 202 the state of the first detection transistor 4 and of the second detection transistor 8 of each pixel 3, so as to verify whether, during the aforesaid exposure step 201, a radiation, in particular a particle, has actually interacted with said pixel 3.

With regard to the exposure step 201, it preferably has, as schematically shown in fig. 4, a time window W _E with a duration between 50 ps and 30 ms.

This window can also be increased at will or decreased outside the range reported herein up to values around a few hundred ns. Finally, it is noted that the time window W _E does not necessarily define the acquisition time of the particle, which can be extrapolated by means of higher-resolution peripheral circuits not described in the present invention.

Furthermore, it will be important, for the reasons indicated below during the description of the operation of the radiation sensor 1 of the invention, to also define a time window W _T with a predetermined duration of the entire acquisition cycle comprising the aforesaid three operating steps.

In particular, according to the preferred embodiment of the invention, said time window W _T with duration of the entire cycle is greater than the duration of the time window W _E by a few microseconds, due to the durations of the reset and reading steps.

As mentioned above, the method for the detection of radiations by means of a solid-state radiation sensor 1 according to the preferred embodiment described above, including all the envisaged variants, is also part of the invention.

In particular, the method of the invention provides for performing, for each pixel 3, in sequence and periodically, a pre-charging step 200, an exposure step 201 and a reading step 202.

The pre-charging step provides for placing in operating mode the first electronic switch 6 and the second electronic switch 10, so as to impose on the first common node A the voltage Vdd and, likewise, so as to impose on the second common node B the voltage Vgnd.

As mentioned above, this makes it possible to establish for the first detection transistor 4 a voltage Vgs between the gain terminal 43 and the source terminal 41 with lower value or, at most, equal to zero.

Furthermore, such a pre-charging step 200 makes it possible to establish for the second detection transistor 8 a voltage Vgs between the gain terminal 83 and the source terminal 81 with greater value or, at most, equal to zero.

These two voltage values then cause a polarization of both the above- mentioned detection transistors 4 and 8 in the cut-off region.

Even more specifically, since, as seen above, the voltage Vdd imposed on the first common node A is higher than the higher voltage Vhigh by a predetermined value AV, the voltage Vgs between the gain terminal 83 and the source terminal 81 of the second detection transistor 8 is positive, while, since the voltage Vgnd imposed on the second common node B is lower than the lower voltage Vlow of the aforesaid value AV, the voltage Vgs between the gain terminal 43 and the source terminal 41 of the first detection transistor 4 is negative, thus placing both MOS capacitors that define the two detection transistors 4 and 8 in the accumulation region.

This aspect, advantageously, allows to completely close the channel of the two detection transistors 4 and 8, preventing in an absolute manner the passage of sub-threshold currents, also known as leakage currents, between the drain and source terminals of each of the aforesaid two detection transistors 4 and 8, which on the contrary could inevitably lead to a change of state of the same detection transistors 4 and 8.

Evidently, as long as the two common nodes A and B are operationally connected respectively with the first supply line 5 of voltage Vdd and with the second supply line 9 of voltage Vgnd, the two detection transistors 4 and 8 are forcibly maintained in a cut-off condition.

However, according to the invention, in order to be able to make such detection transistors 4 and 8 capable of detecting the incidence of a radiation, or of a particle, the method of the invention provides for placing in cut-off mode both the first electronic switch 6 and the second electronic switch 10, so as to make both the first common node A and the second common node B float.

At this point, these common nodes A and B, which are physically defined by the junctions p-n of the two detection transistors 4 and 8, are sensitive to radiations, in particular to particles, which possibly impact against these sensitive parts of each pixel 3.

In fact, in the event that a particle impacts, during this exposure step, the sensitive part of at least one of the two detection transistors 4 and 8, for example the first detection transistor 4, this impact generates charges at the junction p-n of the same detection transistor 4, which in turn cause a current spike that crosses the junction p-n. Since as seen above, at both the first common node A and the second common node B, the two detection transistors 4 and 8 are connected in feedback, this current spike will be imposed at the input, i.e. at the gate terminal of the other detection transistor, for example at the gate terminal 83 of the second detection transistor 8. Accordingly, such a current spike, if it has adequate amplitude and duration, will also be able to place such a second detection transistor 8 in the operation region. In turn, then, the second detection transistor 8 will impose at input to the first detection transistor 4, i.e. at the gate 43 of the first detection transistor 4, a current sufficient to maintain or bring back into operating mode also said first detection transistor 4.

This then results in a change of state of both such detection transistors 4 and 8 that will be detected at the end of the aforesaid exposure step 201, during the aforesaid reading step 202.

This phenomenon is evidently triggered, as mentioned above, when the magnitude of the charges generated by the incidence of a radiation on at least one of the two detection transistors 4 and 8 is sufficient to bring the voltage Vgs between gate and source, for example of the second detection transistor 8, beyond the threshold value.

Furthermore, since, as mentioned above, the voltage Vgs imposed on both the detection transistors 4 and 8 in the pre-charging step 200 is such that the same detection transistors 4 and 8 are in the accumulation region, any triggering of spurious spikes due to the sub-threshold currents is avoided.

At the end of the aforesaid time window W _E of the exposure step 201, as just mentioned, a step 202 of reading the states of the two detection transistors 4 and 8 for each pixel 3 is performed. In particular, the signal read corresponds to a binary signal, i.e. to the value “0” in the event that both the detection transistors 4 and 8 are cut-off and to the value “1” in the event that the aforesaid detection transistors 4 and 8 are in operating mode, i.e. in the event that a sufficiently intense radiation has impacted against the specific pixel 3.

With regard to the step 202 of reading the states of the two detection transistors 4 and 8 for each pixel 3 of the radiation sensor 1 , it could envisage, according to a preferred implementing embodiment, scanning the various pixels 3. The logic signal of each pixel 3 is discriminated by thresholding starting from the value set on the aforesaid supply lines 5 and 9. These supply lines 5 and 9 are pre-charged to a default value before each reading. Once selected, each pixel 3 writes its state on the aforesaid supply lines 5 and 9, while a comparator (which could simply be made by an inverter or a “sense amplifier”) discriminates and amplifies the signal that will then be sampled by means of a local memory on each column at the end of the operation. The stored data will then be sent by means of a serial or parallel reading organized into groups of 8 or more bits. Since the information is differential, the reading can take place in a differential way in order to strengthen the procedure itself. However, alternatively, only the value written in the second supply line 9 could be used. Finally, it is not excluded that reading techniques and circuit implementations that are completely similar to the ones used for the memories could be used.

At the end of this reading step 202, as mentioned above, the method of the invention provides for repeating the three steps cyclically and then, as a first step, provides for performing the aforesaid pre-charging step 200 again.

In this regard, as mentioned above, it is also important to appropriately define the duration of the time window W _T of each cycle, as the shorter this duration is and the less the probability will be of spurious current spikes being triggered on one or more pixels 3, due to for example so-called “dark” events such as the above-mentioned sub-threshold currents or leakage. This, together with the fact of setting in the pre-charging step 200 a negative voltage Vgs for the first detection transistor 4 and a positive voltage for the second detection transistor 8, advantageously allows to raise the detection factor of events triggered by real radiations that affects the pixel, compared to the triggering of events due to such spurious spikes.

From experiments carried out by the applicants, it was found, as understandable from the graph of fig. 5, that, by keeping the value of the higher reference voltage Vhigh constant within 1 ,2V and acting on the value of AV, it is possible to define the amount of critical charge Qcrit that is sufficient and necessary to trigger an event in each of the pixels 3 that make up the radiation sensor 1 of the invention.

In particular, it can be observed that, by reducing the value of AV below 100 mV, it is possible to increase the sensitivity of each pixel 3 up to a critical charge value Qcrit around 1 ke-

From what has been said and illustrated, it can be therefore understood that the radiation sensor of the invention and the method for the detection of radiations achieve all the purposes that have been set.

In particular, the object of realizing a sensor of radiations, in particular of particles, with the sensitive elements (pixels) having a much smaller size than the sensitive elements of the radiation sensors of the known type is achieved. Thus, the object of realizing a sensor of radiations, in particular of particles, having a higher spatial resolution than the radiation sensors of the prior art is also achieved.

A further object to realize a purely digital sensor is also achieved. This, as a first effect, allows the reduction of complexity in terms of the number of transistors and controls.

In addition, the object of realizing a radiation sensor capable of maximising, in perspective, thanks to the digital approach, the technological scalability of its constituent components is also achieved.

Furthermore, the object of realizing a radiation sensor that outputs an amount of reduced information of a digital type, which therefore do not need conversion and that are quickly processable and storable, while maintaining a high spatial and temporal resolution is also achieved.

Finally, the object of realizing a radiation sensor with reduced consumptions is also achieved.