WO2010010519A1

WO2010010519A1 - Time-measurement device for applications without power source

Info

Publication number: WO2010010519A1
Application number: PCT/IB2009/053175
Authority: WO
Inventors: Peter Jan Slikkerveer; Jozef Thomas Martinus Van Beek; Pawel Musial
Original assignee: Nxp B.V.
Priority date: 2008-07-23
Filing date: 2009-07-22
Publication date: 2010-01-28
Also published as: CN102099753A; EP2318889A1

Abstract

The present invention relates to a time-measurement device comprising a timing-reference circuit on a single substrate or on multiple substrates; a source of radioactive radiation on or in the substrate; a counter for radioactive decay events on or in the substrate. A control circuit on the substrate is configured to control the integration time of the counter using the timing reference, receive the count signal from the counter, and calculate a time difference between a momentaneous time and a reference time using the count signal, the value of the integration time span, the half life, and a known decay rate of the at least one radioisotope at the reference time.

Description

TIME-MEASUREMENT DEVICE FOR APPLICATIONS WITHOUT POWER

SOURCE

FIELD OF THE INVENTION

The present invention relates to a time-measurement device, a real-time clock and to an electric component comprising a real-time clock circuit. The invention further relates to an card-shaped identification or authentication device and to a method for measuring time with respect to a reference time.

BACKGROUND OF THE INVENTION

The use of radioactive decay for purposes of measuring time in integrated circuits is known in the art, for instance from US 4,676,661. The time measurement device disclosed in US 4,676,661 aims at replacing quartz-oscillator-based time measurement by counting radioactive decays up to a preset number of decay counts to periodically provide electrical signals, for counting time in units of seconds. Whereas US4676661 also considers to calculate lapsed time since manufacture in the embodiment of Fig. 5, this calculation of time is performed with an accuracy of years only.

Time measurement has received attention for incorporation into many application devices other than watches. However, many of these application devices, such as RFID tags and identification cards, bank cards and the like are usually not continuously connected to a power source. Some of these application devices may allow the incorporation of a battery or of an accumulator, but do not allow or require a continuous operation under consumption of energy from batteries or accumulators when disconnected from power lines. For these application devices, it would be desirable to be able to keep precise track of the real time even in absence of a connection to an electrical power source over a long time, in some cases even in terms of years. SUMMARY OF THE INVENTION

According to a first aspect of the invention, a time-measurement device is provided, comprising: a timing-reference circuit, which is configured to generate an electrical reference signal with a reference frequency suitable for measuring or controlling an integration time span with a resolution of seconds or less; a source of radioactive radiation on or integrated into a substrate and comprising at least one radioisotope with a known half life of at least one month; a detector for radioactive radiation, which detector is integrated with the source of radioactive radiation on or into the same substrate and configured to provide a detector signal indicative of a number of radioactive decay events detected during an integration time span; a control circuit, which is connected with the timing-reference circuit and with the detector and which is configured to - either control or measure the integration time span, and calculate a time difference between a momentaneous time and a reference time using the detector signal, the value of the integration time span, the half life, and a known decay rate of the at least one radioisotope at the reference time.

The time-measurement device of the first aspect of the invention allows keeping track of the time elapsed with respect to the reference time even after a long period of disconnection from power.

To this end, embodiments of the time-measurement device of the present invention comprises a timing-reference circuit, a source of radioactive radiation, an associated detector for radioactive decay events, for instance a counter for radioactive decay events, and the control circuit.

Using the detector signal and known parameters such as a stored reference time, the value of the integration time span, the half life, and a known decay rate of the at least one radioisotope at the reference time, a time difference between a momentaneous time and the reference time can be calculated with an accuracy that provides valuable real-time information even after years of disconnection from power.

In the following, additional features of embodiments of the time-measurement device of the first aspect of the invention will be described. The additional features of different embodiments can be combined with each other to form further embodiments, unless additional features of certain embodiments are described as forming mutual alternatives to each other.

The time-measurement device according to the first aspect of the invention can be provided on a substrate with a thickness of less than 300 micrometer. Suitably, the thickness is even lower for an improved integration into flat application devices such as such as RFID tags, or card-shaped devices, e.g., identification cards, bank cards, credit cards, and the like. A preferred substrate thickness for such applications is less than 150 micrometer. However, the substrate can be thinned even further to thicknesses below 100 micrometer with appropriate care during fabrication. To achieve a particularly small size, the time-reference circuit of preferred embodiments comprises a micro-electromechanical resonator, hereinafter MEMS resonator, and an oscillator circuit, which is connected with the MEMS resonator. In these embodiments, a monolithic integration can even be achieved.

Accordingly, the time-measurement device of the first aspect of the invention is in one embodiment integrated using a single substrate, and thus integrate on the substrate not only the timing-reference circuit, but also the source of radioactive radiation, the counter and the control circuit. This way, a very small device is achieved, which is suitable for incorporation into thin application devices such as those mentioned before, and even for incorporation into paper sheets. Monolithic integration is not an absolute requirement. A small size, in particular with respect to the product height, can also be achieved by providing separate dies, which are combined into a single package or assembly. For instance, a MEMS resonator on one die, a beta-emitter-and-counter die, and a circuitry die including, e.g., driving circuitry, non-volatile memory and logic, can be combined in a single package or assembly, e.g., using die stacking. Such a package is small enough to allow incorporation into smart-cards etc.

However, the time-measurement device of the present invention can also be used in application devices, which do not have such strict requirements regarding geometrical extension. In such embodiments with less restrictions regarding size, the timing-reference circuit may alternatively comprise a quartz oscillator. The time-measurement device of the invention is thus useful also for any application device, which benefits from the ability of keeping track of elapsed time also after the device has been disconnected from its power source.

In one preferred embodiment, the half life of the radioisotope is at least one year. In view of a limited lifetime of application devices, in which the time-measurement device of the present invention is to be used, a radioactive isotope is used in one embodiment, which has a half life of 5 and 10, in another embodiment of between 10 and 20 years. Preferably, the radioisotope used is an emitter of beta radiation, i.e. electrons. Beta emitters are preferred over alpha or gamma emitters due to their generally lower environmental risk. Also, emitted electrons can have a lower energy and, consequently, a lower penetration depth that keeps the radiation from escaping the time-measurement device. Therefore, low-energy emitters are strongly preferred over higher-energy emitters so as to keep the risk caused by the radioactive radiation as close to zero as possible. For a further reduction of the risk of human exposure to emitted radiation, a shielding layer may be provided on the substrate or a shielding may be incorporated into in an application device that contains the time- measurement device.

In one embodiment, a single radioisotope is used as the source of radioactive radiation on or in the substrate. A suitable example is the tritium isotope, ³H. The tritium isotope is a beta emitter with a half life of approximately 12 years. Emitted electrons have an energy of 18 keV. Tritium has the advantage of being implantable into a defined substrate area at a desired implantation dose. Therefore, the number of tritium atoms in the substrate can be controlled with high accuracy according to the requirements of clock operation.

If only a single radioisotope is used as the source of radioactive radiation, the control circuit is preferably configured to calculate the time difference between the momentaneous time and the stored reference time according to the formula

In A₀T - In C(t, T) λ wherein t denotes the time difference, Ag the known decay rate of the radioisotope in the source of radioactive radiation at the reference time, C(t, T) a number of counts detected at the point in time defined by t over the integration time span T, and λ is related to the half life of the radioisotope by the relation λ=0.7 tm.

As will be shown further below in the context of the description of the Figures, the above formula can be derived from the well-known exponential decay law describing the rate of radioactive decay as a function of time. The decay rate Ag of the radioisotope in the source of radioactive radiation at the reference time can be determined by calculation from the implantation dose. It can also be determined by measurement. The measurement of Ag need not be performed at the fabrication site, but can be performed as an initialization procedure at a later point in time, using the counter integrated into the time-measurement device. The measured decay rate A o is to be stored in a non- volatile memory of the time- measurement device.

The integration time span is in preferred embodiments of the order of seconds or less. This allows a quick time measurement will sufficient accuracy for many application cases and keeps power requirements for a time measurement low. With increasing integration time span, on the other hand, a higher accuracy of the time measurement can be achieved. It may be desired for some applications to allow a renewal of the reference time and the associated decay rate at a later point in time. For instance, the time-measurement device could be recycled and used in a second product. In that case, a new reference time is to be set. Note that along with the reference time, the decay rate Ag must be renewed to allow time measurements with respect to the new reference time.

However, in some applications such reprogramming of the reference time is not desired. For instance, the time-measurement device can be used for monitoring of the age of products. For such applications, writing access to the non-volatile memory should be blocked employing technical means, which as such are known in the art. Or a one-time- programmable (OTP) non-volatile memory can be used in the time-measurement device for storing the reference time and the associated decay rate.

In a preferred embodiment, the substrate is a semiconductor substrate that includes a buried layer containing the radioisotope, wherein the detector for radioactive decay events comprises a pn-junction with a detector zone of the substrate, which detector zone is depletable from charge-carriers at an operating voltage, and wherein the buried layer at least partially extends within the detector zone. The detector zone is thus formed by a carrier-depleted zone associated with the pn-junction.

This way, it is achieved that the source of radioactive radiation is placed in close vicinity to the detector, which can be realized as a depletion region of a CMOS diode structure. The radioactive emission that is absorbed in the depletion region creates electron- hole pairs. The electrons are pulled to the n-conductive region, while holes are pulled to the p-conductive region, as a result of the electric field that is present inside the depletion region. Electron-hole pairs generated while the time-measurement device is not connected with a power source, may at least partly recombine due to their limited lifetime. A charge build-up during power-off might form an initial current pulse when reconnecting the device to power, which can be neglected by proper operation of the device. For instance, the integration time can be started with a suitable delay to avoid artificial counts from decay events that have occurred during power off. Therefore, the measurement of the number of counts performed to determine the point in time with respect to the reference time is not disturbed by false counts, which are associated with decay events that do not fall into the integration time span. This way, an accurate measurement of the number of decay events during the integration time span is ensured, resulting in a precise and reliable determination of the time difference with respect to the reference time.

As an alternative to the use of a single radioisotope, the source of radioactive radiation may comprise two radioisotopes, i.e., a first radioisotope and a second radioisotope. In this alternative embodiment, a second half life of the second radioisotope is preferably at least three times larger than a first half life of the first radioisotope, which first half life preferably ranges between 10 and 20 years. A disadvantage of this embodiment is that the second radioisotope will not have been decayed to a large extent at the end of a typical product lifetime. For this reason, the use of a single radioisotope is currently preferred. However, if two radioisotopes are to be used, they should preferably be provided in numbers such that decay rates of the two radioisotopes, as measured in decays per second are identical or at least approximately identical at a know point in time before first operating the time-measurement device. This known point in time could for instance be the reference time. Using this embodiment, the control circuit is preferably configured to calculate the time difference according to the following formula: In C₁ - In C₂ + In A₀₂ - In A₀₁

X₂ - X_x wherein t denotes the time difference, Aoi and A02 the known decay rates of the first and second radioisotopes in the source of radioactive radiation at the reference time, Ci(t, T) and C2(t, T) respective numbers of counts detected at the point in time defined by t over the integration time span T from the first and second radioisotopes, and λi and λ₂ are related to the respective half lifes tm of the first and second radioisotopes by the relation X=O.7 tm. The above formula is similar to that provided for the use of a single radioisotope, and can also be derived starting from the known exponential decay law.

Note that, for simplicity of reference, λ, λi and λ₂ will in the present application sometimes be referred to as the respective half lifes of the subject radioisotope(s), notwithstanding the validity of the relation X=O.7 tm. In a time-measurement device that uses two radioisotopes, the detector is preferably configured to differentiate between counts of at least two energy ranges associated with decay events, for separating first decay events of the first radioisotope from second decay events of the second radioisotope, and to separately count the first and second decay events.

As in embodiments that use only one radioisotope, a semiconductor substrate may include the detector and a buried layer containing the first radioisotope. In contrast to embodiments with only one radioisotope, a layer consisting of or comprising the second radioisotope may suitably be arranged on an insulating layer on the substrate.

One-time-programmable or other memory means, which are not reprogrammable, may be provided in the time-measurement device for storing the known half life of the radioisotopes. However, the provision of such memory means for storing the half- life value is not a requirement. The half- life value may be implemented in hardwired form in a corresponding multiplier, performing a multiplication by 1/λ, or 1/(X₂-X₁) according to one of the formulas mentioned above, depending on whether one or two radioisotopes are used in the embodiment.

The used reference time can have the value "zero" in units of, e.g., seconds, minutes, or days. In this case, no programmable non- volatile memory is required for storing the reference time.

The time-measurement device of the present invention is also very well suited for use in a real-time clock. A real-time clock (RTC) is a device that keeps track of the current time. A real-time clock comprising a time-measurement circuit of the first aspect of the invention or one of its embodiments described herein uses a real-time value as the reference time. For instance, instead of using a reference time "0 seconds", a real-time given for instance by date (day, month, year) and time of day (hour, minute, second) is used. The accuracy of the reference time can be chosen according to the needs of a particular application. A current or, in other words, momentaneous real-time value, or a corresponding signal indicative of such a real-time value, is provided as an output by the real-time clock.

In one embodiment of the real-time clock, the control circuit is to this end additionally configured to calculate and provide at its output a real-time value using the stored reference time and the calculated time difference between the momentaneous time and the reference time. As explained earlier, the momentaneous time is the time, at which the measurement of the radioactive decay rate is performed.

The real-time clock thus shares the advantages of the time-measurement device of the invention. It provides accurate real-time information even after years of disconnection from power. It can be implemented with particularly small size suitable for use in cards, RFID tags, and the like. In these and other application cases, a programmable, non-volatile memory forms a useful part of the time-measurement device. This will allow setting the reference time according to the needs of a particular application, and determining a real-time value, which can be output in terms of current day, month, year, hour, minute, and second using the stored real-time value reference time and the measured time difference. Of course, the exact format of the real-time output should be selected with a view to the precision of the time measurement. A detailed discussion regarding the precision of the time measurement is given further below in the context of the description of an embodiment with reference to the enclosed Figures. In other useful embodiments, a programmable, non- volatile memory can additionally or alternatively be used for other purposes. For instance, it can be used for storing a maximum lifetime of a product, to which the time-measurement device is associated. As an illustrative example, a packaging of perishable products may comprise a time-measurement device according to an embodiment of the present invention. In the non- volatile memory of this time-measurement device, not only the reference time may be stored, but also the maximum lifetime, as calculated from the reference time. This way, the product can be checked for having passed its lifetime without having to refer to other means such as a database. In other application contexts, such as identification and authentication, the programmable non-volatile memory can be used for storing a password and/or other data required in this context, such as a name or other identification data.

An electric component that comprises a time-measurement device according to the first aspect of the invention or one of its embodiments, or a real-time clock according to the mentioned embodiment, forms a further aspect of the invention. Examples of electric components are integrated circuits, assemblies of integrated circuits, tags, and application products incorporating the time-measurement device or the RTC. Examples of such application products are cards used for identification or authentication, sheets of card board, and sheets of paper.

The electric component preferably comprises a power terminal, which is configured to receive power from an external power source, for operation of the electric component, at least for operation of the timing-reference circuit, the counter, and the control circuit. Depending on the power transfer mechanism employed, additional means may be provided for transforming the received power and for providing an operating voltage to be applied to the real-time clock in the event of receiving electrical power from the external power source. For certain applications, such as RFID tags, the power terminal is configured to receive power via a wireless power transfer. In such embodiments, the electric component preferably has a transmitter, which is connected with the control unit and configured to transmit the calculated time difference or the calculated real-time value via a wireless communication channel in the event of receiving power. This way, a fully wireless operation of the electric component is enabled.

In another embodiment, the power terminal is configured for connection to a battery. The time-measurement device of the present invention allows a considerable power saving over known time-measurement devices, since power is not required continuously, but only during short intervals over time, namely, in the event of a time measurement. Therefore, the lifetime of a battery can be extended considerably, reaching the range of a typical projected product lifetime for many products.

Note that some of the functions of the time-measurement device may be provided either in hardware or in the form of code executable by a microprocessor. For instance, the functionality of the control unit provided by the time-measurement device of the first aspect of the inventions can be implemented using a programmable microprocessor.

In an independent second aspect of the invention, a method for measuring time with respect to a reference time is provided. The method of the second aspect of the invention comprises - providing a time-measurement device according to the first aspect of the inventions or one of its embodiments, which may in particular be a real-time clock; keeping the time-measurement device or real-time clock disconnected from a power source for a time span; providing operating power to the time-measurement device or real-time clock from a power source after the time span; letting the control unit of the time-measurement device or the real-time clock a) measure or control an integration time span b) receive the detector signal from the detector, and c) calculate a time difference between a momentaneous time and a reference time using the count signal, the value of the integration time span, the half life, and a known decay rate of the at least one radioisotope at the reference time; d) optionally and only if a real-time clock according to claim 11 is provided, additionally calculate a real-time value using the stored reference time and the calculated time difference; and receiving a time-difference value or real-time value from the time- measurement device, respectively, depending whether step d) has been performed.

Any combination of the steps a) to d) may be performed by a microprocessor by means of executable code suitable for performing the respective method steps. By virtue of the use of the time-measurement of the first aspect of the invention, the method of the present aspect allows performing the time measurement even after a long time span of disconnection from a power source. Due to the small size achievable for the time measurement device of the first aspect of the invention, the method is widely applicable in many fields, including product monitoring, identification, authentication, certification (for instance of documents), and more.

Using the method of the present invention it is for instance possible to precisely and automatically determine the age of documents, or of products "on the shelf, without having to look at the document or product or refer to any data base, in which the individual product is stored. Further embodiments are defined in the dependent claims.

The present invention is further elucidated in the following with reference the enclosed Figures, which provide examples, which are not intended to limit the scope of the invention. The person skilled in the art will understand that various embodiments may be combined.

BRIEF DESCRIPTION OF THE DRAWINGS

These and other aspects of the invention will be apparent from and elucidated with reference to the embodiment(s) described hereinafter. In the following drawings

Fig. 1 shows a schematic and simplified block diagram of a first embodiment of a time-measurement device;

Figs. 2 and 3 show a cross-sectional and a top-view, respectively, of a first embodiment of a time-measurement device in a section comprising a source of radioactive radiation and a counter for radioactive decay events;

Fig. 4 shows a cross-sectional view of a second embodiment of a time- measurement device in a section comprising the source of radioactive radiation and the counter for radioactive decay events;

Figs. 5 to 11 show different stages of a manufacturing process for a micro - electromechanical resonator in a time-measurement device according to a third embodiment; Fig. 12 shows a simplified flow diagram of a method for measuring time according to an embodiment of the invention.

DETAILED DESCRIPTION OF EMBODIMENTS Fig. 1 shows a schematic and simplified block diagram of a first embodiment of a time-measurement 100 device.

The time-measurement device 100 of Fig. 1 comprises a timing-reference circuit 102. The timing-reference circuit 102 has a MEMS resonator 104. Structural details of the MEMS resonator will be described for an illustrative example further below. Generally, MEMS resonators as such are known in the art. The MEMS resonator 104 can be fabricated to provide specifications (regarding Q, frequency stability) equal or close to those known from quartz resonators. MEMS resonators can be used in oscillator circuits such as oscillator circuit 106 to provide a frequency stability and with controllable oscillation frequencies in the range of up to the GHz range. An example of a suitable timing-reference circuit connected with a MEMS resonator is for instance disclosed in WO 2008/033681 A2, which is incorporated herein in its entirety. An oscillator circuit is disclosed in Fig. 1 of this document. Paragraph [0006] of WO 2008/033681 A2describes this oscillator circuit that is connected with a MEMS resonator in more detail. As there, the MEMS resonator of the present embodiment may be connected between a drive amplifier and a sense amplifier. A MEMS resonator drive circuit comprised by the oscillator circuit may receive the output of the sense amplifier, which may be a sine wave. The MEMS resonator drive circuit comprised by the oscillator circuit operates on the input signal to produce a drive signal. A typical drive signal is a continuous signal that may be generated by modifying an amplitude of the input sine wave or by clipping the input sine wave to produce a square wave.

Using the MEMS resonator, the oscillator circuit is able to provide an electrical reference signal with a reference frequency, which depends on the oscillation frequency. The reference signal is provided to the control circuit 108.

The timing-reference circuit 102 thus serves to provide a reference signal for a timing operation performed by the control circuit. This timing operation includes the control of the integration time span of a counter 110 for radioactive decay events.

The radioactive decay events are provided by a source of radioactive radiation, which is symbolized in Fig. 1 by the lower-case Greek letter β. In the present embodiment, the source of radioactive radiation β emits β-radiation, i.e. electrons. The counter is configured to provide a count signal indicative of a number of radioactive decay events detected during an integration time span.

The integration time span is controllable via a start signal and a stop signal, which may be generated and provide by the control circuit 108. For controlling the exact duration of the integration time span, the control unit 108 uses the reference signal provided by the timing-reference circuit 102. The control circuit may use at least one stored control parameter such as a predetermined value of the integration-time span.

In operation, the control circuit may thus provide a start signal for the counter for radioactive decay events and at the same time start a measurement of the integration time span using the reference signal, receive the reference signal from the timing-reference circuit, and provide the stop signal for the counter at the end of the integration time span. Alternatively, the beginning and end of the integration time span may be triggered by an external device, and the control circuit measures the duration of the integration time span.

On the basis of the count signal from the counter 110, the value of the integration time span, the known half life of the source of radioactive radiation and a know decay rate of the radioisotope at a reference time, a time difference between a current (i.e., momentaneous) time and the reference time is calculated by the control unit 108.

The formula for calculating the time difference will be derived in the following starting from well-known laws of radioactive decay. In addition, mathematical relations will be derived, which allow determining suitable parameter ranges of the time- measurement device, a) General considerations and approximations

Radioactive decay, such as β (beta) decay, of a particular radioisotope follows an exponential decay law as a function of time t: A(t) = A₀ exp(-λt) (eq. 1)

Here, A denotes a decay rate, i.e., a number of decays per unit of time, A₀ is a number of decays at t=0, and λ=0.7/tm , where tm is the half life of the isotope.

Detectors for radioactive decay events allow counting the number of decay events in a certain time span, the integration time. The integration time of a detector is denoted T. Assuming that the integration time is much shorter than the half life, T«\lλ, the number of counts detected is given can be determined by an approximation to the integral of eq. 1 between a starting time of the integration and an integration time span T after that. This approximation can be visualized as a rectangle under the exponential-decay curve, the rectangle having a width T on the time axis and a height corresponding to the decay rate given by (eq. 1) at the beginning of the integration time. Assuming, by way of example, the beginning of the integration time at t=0, the count number C(T) is

C(t = 0,T) = A₀T . (eq. 2) For any later measurement time starting at a time t and having an integration time T one finds the integrated counts at the end of the measurement (integration time), i.e., at t+ T to be

C(t,T) = A₀Tex_V(-λt) . (eq. 3) b) Calculating elapsed time using only one radioisotope When only one radioisotope is present, eq. 3 can be used to determine an elapsed time as

, =^ W. (eq. 4) c) Accuracy of time measurement for the case of only one radioisotope Taking the C and T partial derivatives of eq. 4 results in dt 1

BT ^~ XT

The standard deviation in the number of counts sets the time accuracy of the detector. The standard deviation in time is related to the standard deviation in the number of counts. The standard deviation in C is given by, σ_c = Vc . (eq. 6) Combining eqs. 3 to 6 allows deriving an expression standard deviation, and thus for the accuracy of measuring the time t as a function of AQ , T and t, dt dt 1 ( λt λ 1 σ _r r _π\ σ , = — - σ _c + —σ_τ = -^==exp — + --^ . (eq. 7) λ jAJ V 2 ) λ T

The first term gives the uncertainty in the determined value of the elapsed time as a result of the decay statistics. It can be seen that the uncertainty increases with elapsed time, and decreases with increasing initial decay rate (corresponding to the number of radioisotopes provided), and with the integration time. The second term gives the uncertainty a result of the accuracy of the reference oscillator. This contribution decreases with increasing integration time. As an application example, from eq. 7 one obtains for the radioisotope ³H with ti/2= 12 yr (= years), AO= 2J3E+08 s-1, t =10 yr, and an integration time span of T= lsec the requirement σ_τ/T < l60 ppm (eq. 8) when the uncertainty is set to σ_t=l day. This value can be achieved using state-of-the-art MEMS resonators, d) Calculating elapsed time using two radioisotopes

In the following, subscript indices 1 and 2 used with the symbols of the physical quantities defined above refer to first and second radioisotope used. From eq. 1 one obtains the ratio of the number of decays for two isotope, C\ and C₂, and dividing the number of counts for both isotopes gives

^ = ^cx_V((λ₂ -λ_ι )t) (eq. 9)

¹CG

Eq. 8 can now be used to determine elapsed time, ^ In C₁ - In C₂ + In A₀₂ - In A₀₁ λ, - A e) Accuracy of time measurement for the case of using two radioisotopes Taking the C derivative of eq. 10 gives dt _ 1 3C₁ ^" (A₂ - A₁ )C₁

(eq. 11a and l ib) dt 1

SC₂ (A₂ - A₁ )C₂

Combining equations 6, 9, and 11 gives an expression for the accuracy of measuring t as a function of ^₀₁ , A₀2 , Ai , A₂ ,and t,

12)

From eq. 12 it can be concluded that the isotope with the lowest activity AQ determines uncertainty in elapsed time, σ_t. Furthermore, it can be seen that σ_t is infinite when Ai =A₂. Therefore, the smallest uncertainty in σ_t at lowest levels of activity AQ is obtained, when

f) Selection of suitable radioisotopes for the case of two radioisotopes Combining eqs. 12 and 13 gives,

tmax denotes a desired maximum of the elapsed time that is to be measured. In other words, it corresponds to a projected product life time. Setting the partial derivative d I dλ \

according to eq. 14 equal to zero allows determining a value for λ_\ that requires the smallest amount of AQ_\ (and A ) for a given integration time T,

2

K =- (eq. 15)

Therefore, in order to minimize Am and Ao₂, the shortest lived isotope should have a half life close to t₁/2=1.4t_max. For the product life time of 10 yr this translates to ti/2=14yr, which is close to the half life of ³H, ti/₂= 12 yr. Choosing ⁶³Ni with ti/₂= 92 yr as the second isotope fulfils the condition λ₂ »λi. g) Determining a suitable number of nuclei of the radioisotopes in the device

Using eq. 14 it is now possible to calculate the minimum activity Aoi and Ao₂ for both isotopes and the corresponding number of nuclei, No at t=0. The number of required radioactive nuclei, No at ^=O is related to Ag through,

N₀ = - Λ . (eq. 16) 0 λ h) Requirements for counter and chip area used

From the numbers given in Table 1 and characterizing the two radioisotopes ³H and ⁶³Ni, an indication of required processing capabilities and of the capability of the read-out electronics can be derived.

Table 1 Parameters of radioisotopes ³H and ⁶³Ni

For the H based detector, No combined with the maximum realizable implantation dose (set by the implanter equipment and H source) sets the minimum required chip area. The Ag sets the speed of the electronic counter. In this case the counter needs to be able to record pulse rates of at least 270 MHz.

From Table 1 can be seen that the number of radioactive nuclei for ⁶³Ni is approximately 1OX the number needed for H. This can be accommodated by an appropriate thickness of a Ni layer and/or by appropriately large lateral extensions of the Ni layer.

It should be taken into consideration that, because of the long half life Of⁶³Ni, this radioisotope will still decay when the time-measurement device or the product it is used in is no longer in use. Therefore, it is currently preferred to use only one (short-lived) radioisotope in the device. Next, examples for structures of the counter and the MEMS resonator will be given with reference to Figs. 2 and following.

Figs. 2 and 3 show a cross-sectional and a top-view, respectively, of a first embodiment of a time-measurement device in a section comprising a source of radioactive radiation and a counter for radioactive decay events. The time-measurement device 100 of Figs. 2 and 3 is integrated on a single substrate. The counter 110 is implemented in a single wafer that also comprises the timing-reference circuit 102 with the MEMS resonator 104 and the oscillator circuit 106, the control circuit 108 and the counter 110. As mentioned before, this is an advantageous option, not an absolute requirement. The section of the substrate 112 shown in Figs. 2 and 3 only shows the counter 110 and the source β of radioactive radiation. The counter comprises a CMOS diode structure with depletion regions 114 and 116 formed close to a junction between p-doped regions 118 and 120 on one hand and n-doped regions 122, 124 and 126 on the other hand. The substrate 112 in this embodiment is a silicon substrate, which is n-doped. The geometrical shape and the dopant concentration provided in the respective n- and p-type zones 118 to 126 determines the exact extension and shape of the depleted zones 114 and 116. The source of radioactive radiation is formed by a buried layer β containing tritium atoms. Electrodes 128 and 130 are connected with the p-regions 118 and 120, or with the n-regions 122 to 126, respectively.

In operation of the counter 110, an operating voltage is applied to the diode structure. The operating voltage is applied to drive the pn-junctions in a reverse bias. The operating voltage determines the size of the depletion zones 114 and 116. β emission (i.e., electrons) that is generated and absorbed in the depletion regions creates electron hole pairs. Due to the electric field present in the depletion zones, electrons are pulled to the n-regions 122 to 126, while holes are driven to the p-regions 118 and 120. The absorbed β-particles results in a current pulse that is detected by a current sensor (not shown) and counted as a single decay event.

A comb-shaped arrangement of the n⁺ and p⁺ regions 118 to 126 is used to maximize the area of the depletion region and therefore increase the sensitivity of the detector.

Fig. 4 shows a variant of the counter 110 for an embodiment that uses two different radioisotopes. The structure of the counter 210 of Fig. 4 resembles that of the counter 110 shown in Fig. 2. The cross-sectional view of Fig. 4 reveals as one difference a surface layer Of⁶³Ni with about the same lateral extensions as the buried layer β. For distinction from the previous embodiment, the buried layer comprising ³H atoms is labeled βl in Fig. 4, and the ⁶³Ni layer is labeled β2. As before, ³H corresponds to the first radioisotope and ⁶³Ni corresponds to the second radioisotope.

The operational principle of the counter for the structure of Fig. 4 is similar to that described for the embodiment of Figs. 2 and 3. A difference to the previous embodiment is that the current sensor (not shown) of the counter is configured to distinguish current pulses according to the total amount of charge transported by the current pulses. The background of this differentiation is that ⁶³N emits β-particles with an energy of 70 keV, which gives rise to a larger number of electron-hole pairs generated per decay event in the detector 210 in comparison with current pulses generated by β particles emitted from ³H, which have an energy of 18 keV. Using the current pulse differentiation, which as such is known in the art, count rates for β-emission from ⁶³Ni and ³H can be obtained concurrently.

Figs. 5 to 11 show different stages of a manufacturing process for a micro - electromechanical resonator in a time-measurement device according to a third embodiment. The Figures show schematic views of an illustrative example of a MEMS resonator, which can be used in the time-measurement device 100 of Fig. 1. The MEMS structure can be integrated on the substrate level in a CMOS front-end process. As an alternative, the MEMS resonator can be integrated into an interconnect stack on different metal levels using known back-end processing techniques. The following description will be given without referring to a particular one of these two alternatives. The described processing is valid for both of them. On a first layer 302, a sacrificial layer 304 is arranged, followed by a second layer 306. The first layer 302 serves as a bottom layer, which in the MEMS resonator may form a bottom electrode. It is made of an electrically conductive material, such as Cu.

The sacrificial layer 304 is partly removed through an opening 308 fabricated in the second layer 306 by conventional patterning techniques. A gap 310 created by this process is subsequently filled by a second sacrificial layer 312, which does not have to fill the gap 310, but covers a lateral section of the second layer 306. A capping layer 314 is then deposited and patterned to comprise openings 316, only one of which is labeled in Fig. 9. The second sacrificial layer 312 is then removed through the openings 316, leaving a cavity 318, in which a resonating element 320 is suspended. Since the resonating element 320 is made of an electrically conductive material, namely, the material of the second layer 306, the motion of the suspended resonating element 320 can be controlled by controlling a voltage applied between the first layer 302 and the resonating element 320. An oscillating motion of the resonating element can be triggered by applying a AC control voltage. Further details of stabilizing the motion of the resonating element at its resonance frequency have been described with reference to Fig. 1.

The cavity 318 is sealed by a sealing layer 322. Note that the structure of the MEMS resonator 300 is only an illustrative example. A wealth of different MEMS resonators is known in the art. As can be seen, both the resonator and the package can be manufactured created using known CMOS processing technology. The thickness of the packaged resonator is about 10 μm. The substrate only serves as a mechanical carrier for the packaged resonator and is typically much thicker than the packaged resonator. Standard substrate thinning techniques, such as grinding and etching, can be used to thin down the substrate. For Si wafers a substrate thickness of less than 150 μm can be achieved. This allows the integration into thin objects such as sheets of paper and smart card format e-passports.

Fig. 12 shows a simplified flow diagram of a method for measuring time with respect to a reference time according to an embodiment of the invention.

The method is based on the use of a time-measurement device according to the first aspect of the invention or of a real-time clock, such as the time-measurement device of Fig. 1. After keeping the time-measurement device or real-time clock disconnected from power for some time (any time span up to a number years within the half life of the radioisotope), the time-measurement is started (step SlO) by providing power to the device. This will trigger the control unit to provide the start signal for the counter for radioactive decay events (S 12) and at the same time start a measurement of the integration time span using the reference signal (step S 14). For the measurement of the integration time span T, the reference signal is received from the timing-reference circuit (step S 16). With reaching the integration time span T, the stop signal is generated by the control unit and provided to the counter (S18). The counter in turn provides the count signal, i.e., the number of counts detected during the integration span to the control unit (step S20). From that, the control unit calculates a time difference between a momentaneous time and a reference time using the count signal, the value of the integration time span, the half life, and a known decay rate of the at least one radioisotope at the reference time (S22). This time difference can be further processed in step S22 to provide a real-time value in "human" units of year, day, and time of day. The method can be summarized as follows:

SlO Start

S 12 Provide the start signal for the counter for radioactive decay events

S 14 At the same time, start a measurement of the integration time span using the reference signal,

S 16 Receive the reference signal from the timing-reference circuit, S18 Provide the stop signal for the counter at the end of the integration time span, S20 Receive a count signal from the counter, and S22 Calculate a time difference between a momentaneous time and a reference time using the count signal, the value of the integration time span, the half life, and a known decay rate of the at least one radioisotope at the reference time. S24 Stop

While the invention has been illustrated and described in detail in the drawings and foregoing description, such illustration and description are to be considered illustrative or exemplary and not restrictive; the invention is not limited to the disclosed embodiments.

Other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality. A single unit may fulfill the functions of several items recited in the claims. Also, several different units may be provided to fulfill different functions of the control unit recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measured cannot be used to advantage.

Any reference signs in the claims should not be construed as limiting the scope.

Claims

CLAIMS:

1. A time-measurement device, comprising a timing-reference circuit, which is configured to generate an electrical reference signal with a reference frequency suitable for measuring or controlling an integration time span with a resolution of seconds or less; - a source of radioactive radiation on or integrated into a substrate and comprising at least one radioisotope with a known half life of at least 1 month; a detector for radioactive radiation, which detector is integrated with the source of radioactive radiation on or into the same substrate and configured to provide a detector signal indicative of a number of radioactive decay events detected during an integration time span; a control circuit, which is connected with the timing-reference circuit and with the detector and which is configured to:

- either control or measure the integration time span, and

- calculate a time difference between a momentaneous time and a reference time using the detector signal, the value of the integration time span, the half life, and a known decay rate of the at least one radioisotope at the reference time.

2. The time-measurement device of claim 1, wherein the time-reference circuit comprises - a micro-electromechanical resonator, hereinafter MEMS resonator, and an oscillator circuit, which is connected with the MEMS resonator.

3. A time-measurement device according to claim 1 or claim 2, wherein the source of radioactive radiation is a single radioisotope, which is an emitter of beta radiation and has a half life of between 10 and 20 years.

4. A time-measurement device according to claim 3, wherein control circuit is configured to calculate the time difference according to the formula

_ In 47 - In C(J₉T⁷)

* ^~ λ ' wherein t denotes the time difference, Ag the known decay rate of the radioisotope in the source of radioactive radiation at the reference time, C(t, T) a number of counts detected at the point in time defined by t over the integration time span T, and λ is related to the half life of the radioisotope by the relation λ=0.7 tm-

5. A time-measurement device according to one of the previous claims, wherein the detector and a buried layer containing the radioisotope are integrated into a semiconductor substrate, the detector comprises a pn-junction with a detector zone of the semiconductor substrate, which detector zone is depletable from charge-carriers at an operating voltage, and wherein the buried layer at least partially extends within the detector zone.

6. A time-measurement device according to one of the previous claims, wherein - the source of radioactive radiation comprises a first radioisotope and a second radioisotope, and wherein a second half life of the second radioisotope is at least three times larger than a first half life of the first radioisotope, which first half life ranges between 10 and 20 years.

7. A time-measurement device according to claim 6, wherein the number of atoms of the first and second radioisotopes in the source of radioactive radiation are chosen such that decay rates of the two radioisotopes are identical or approximately identical at a known point in time before first operating the time-measurement device.

8. A time-measurement device according to claim 6 or 7, wherein the control circuit is configured to calculate the time difference according to the formula

In C₁ - In C₂ + In A₀₂ - In A₀₁ λ₂ - A₁ wherein t denotes the time difference, AQ_I and A02 the known decay rates of the first and second radioisotopes in the source of radioactive radiation at the referece time, Ci(t, T) and C2(t, T) respective numbers of counts detected at the point in time defined by t over the integration time span T from the first and second radioisotopes, and λi and λ₂ are related to the respective half lifes tm of the first and second radioisotopes by the relation λ=0.7 tm-

9. A time-measurement device according to one of the claims 6 to 8, wherein the detector and a buried layer containing the first radioisotope are integrated into a semiconductor substrate, the counter comprises at least one pn-junction with a detector zone in the semiconductor substrate, which detector zone is depletable from charge-carriers at an operating voltage, the buried layer at least partially extends within the detector zone, layer of the second radioisotope is arranged on an insulating layer on the substrate, and - the counter is configured to differentiate between counts of at least two energy ranges associated with decay events for separating first decay events of the first radioisotope from second decay events of the second radioisotope, and to separately count the first and second decay events.

10. The time-measurement device of one of the previous claims, wherein the substrate has a thickness of less than 200 micrometer.

11. A time measurement device according to one of the previous claims, which either comprises a plurality of dies integrated into a single package or in which the timing- reference circuit, the source of radioactive radiation, the detector, the control circuit, and a programmable non-volatile memory are integrated into a single die.

12. A real-time clock comprising a time-measurement device according to one of the claims 1 to 11, wherein the control circuit is configured to calculate and provide at its output a real-time value using the stored reference time and the calculated time difference.

13. An electric component comprising a time-measurement device according to one of the claims 1 to 11 or real-time clock according to claim 12, and a power terminal, which is configured to receive power for operating the timing-reference circuit, the counter, and the control circuit from an external power source.

14. An identification or authentication device, comprising an electric component according to claim 12 or 13.

15. A method for measuring time with respect to a reference time, comprising providing a time-measurement device according to claim 1 or a real-time clock according to claim 11 keeping the time-measurement device or real-time clock disconnected from a power source for a time span; providing operating power to the time-measurement device or real-time clock from a power source after the time span; letting the control unit of the time-measurement device or the real-time clock a) measure or control an integration time span b) receive the detector signal from the detector, and c) calculate a time difference between a momentaneous time and a reference time using the count signal, the value of the integration time span, the half life, and a known decay rate of the at least one radioisotope at the reference time; d) optionally and only if a real-time clock according to claim 11 is provided, additionally calculate a real-time value using the stored reference time and the calculated time difference; and receiving a time-difference value or real-time value from the time- measurement device, respectively, depending whether step d) has been performed.