DEVICE AND METHOD FOR MONITORING A SEMICONDUCTOR COMPONENT

Description

State of the art

[0001] The invention relates to a device and a method for monitoring a semiconductor device. Dielectric breakdown is always a relevant factor with regard to the lifetime of semiconductor devices consisting of at least two electrodes and at least one intermediate dielectric layer, which are operated under an applied electrical voltage. Another factor limiting the lifetime of the semiconductor device is the development of high leakage currents or the decrease in performance due to destruction of the active area of ​​the device.

[0002] From documents DE 10 2016 109137 B3, US 2019 / 369157 A1, US 6 348 806 B1 and the publication "Leakage Current in Low-Temperature PVD PZT Films" by Reis Daniel Monteiro Diniz et al., a method for monitoring a semiconductor device is known which at least partially comprises features of the preamble of the method for monitoring a semiconductor device according to claim 1. Disclosure of the invention

[0003] The device and method according to the independent claims enable the monitoring of semiconductor devices with dielectric layers during operation, allowing not only the detection of defects but also the prediction of their remaining useful life. In addition to tracking the kinetics up to dielectric breakdown, the device also allows for inferences about the subsequent degradation process. This enables lifetime monitoring that incorporates leakage current levels, first peak breakdown, and / or the performance of the semiconductor device (specifically, polarization), as well as leakage current evolution and performance development.

[0004] The method for monitoring a semiconductor device involves detecting a leakage current flowing through a first and second electrode of the device during operation. This leakage current is then compared to a first limit value, and an output is determined based on the result of this comparison. Alternatively, a time point at which an extremum, particularly a maximum, of the leakage current occurs is determined, and an output is determined based on this time point. This output represents a state of the semiconductor device and is then displayed. This allows for lifetime monitoring of the semiconductor device's internal state.

[0005] According to the invention, it is provided that, depending on the time at which the extremum point, in particular the maximum, of the leakage current occurs, a residual value of a lifetime of the semiconductor device is determined, the output comprising the residual value.

[0006] It may be provided that the leakage current in an operation of the semiconductor device in a first operating mode is detected, whereby the operation of the semiconductor device in the first operating mode continues if the leakage current exceeds the first limit value or if the remaining lifetime is greater than a first threshold value.

[0007] It may be provided that a second limit value is determined from the multitude of limit values, whereby the operation of the semiconductor device continues in a second operating mode or the operation is terminated if the leakage current exceeds the second limit value, or that the operation of the semiconductor device continues in a second operating mode or the operation is terminated if the remaining lifetime is greater than a second threshold value.

[0008] It may be provided that a reference limit and a reference temperature are specified, a current temperature of the semiconductor device or of an environment of the semiconductor device is determined, a factor is determined depending on the current temperature and the reference temperature, the reference limit is scaled by the factor, and the first limit and / or the second limit is determined depending on the reference limit scaled by the factor.

[0009] It may be provided that a text is determined which includes the state of the semiconductor device and an indication of a remaining lifetime, whereby the text is sent in a message or output at a human-machine interface.

[0010] It may be provided that a control signal is determined which specifies an operating mode for the semiconductor device, wherein the control signal is output to drive the semiconductor device or a device comprising the semiconductor device to operate in that operating mode.

[0011] The device for monitoring the semiconductor device according to the invention is defined in claim 7.

[0012] The invention also relates to a micro-electro-mechanical system (MEMS), a memory, an actuator, a micromirror, a printhead or a loudspeaker comprising this device for monitoring a semiconductor device and the semiconductor device.

[0013] Further advantageous embodiments will become apparent from the following description and the drawing. The drawing shows: Fig. 1 a schematic representation of a semiconductor device, Fig. 2 a first state for a first distribution of defects, Fig. 3 a second state for the first distribution of defects, Fig. 4 a third state for the first distribution of defects, Fig. 5 a fourth state for the first distribution of defects, Fig. 6 a fifth state for the first distribution of defects, Fig. 7 a first course of a leakage current level according to the first distribution, Fig. 8a second course of a leakage current level according to the first distribution, Fig. 9 a first state for a second distribution of defects, Fig. 10 a second state for the second distribution of defects, Fig. 11 a third state for the second distribution of defects, Fig. 12 a fourth state for the second distribution of defects, Fig. 13 a fifth state for the second distribution of defects, Fig. 14 a leakage current level profile according to the second distribution, Fig. 15 a device for monitoring the semiconductor component, Fig. 16 a method for monitoring the semiconductor component.

[0014] The following description utilizes knowledge of a failure mechanism in dielectric layers to derive a prediction of the remaining lifetime from measurable changes in a semiconductor device comprising one or more dielectric layers during operation. This prediction can be used in safety-critical applications to issue individual, usage-specific maintenance instructions or warnings.

[0015] Conventional lifetime specifications, maintenance intervals, or active component monitoring are designed to prevent critical failures through early replacement or to indicate a fault when it is already too late and the component has failed. The established maintenance / replacement intervals are statistically determined and often reflect very conservative limits for all parts under an assumed average load profile. Load profiles are crucial but can only be estimated.

[0016] In contrast, it was found according to the invention that such semiconductor devices exhibit a deterministic or predictable degradation profile, characterized by continuous failure or gradual behavior. This knowledge of the behavior allows for specific monitoring of the component. At defined limit values, a warning or a controlled replacement can be scheduled. Maintenance and replacement can thus be determined on a component-specific basis and depending on the actual load profile. This offers an advantage for the customer, who only replaces a component when it is actually necessary.

[0017] The gradual behavior can be, for example, a gradual increase in leakage current or a gradual decrease in power / performance.

[0018] Leakage current can occur in semiconductor devices with a dielectric layer positioned between two electrodes. The leakage current profile changes, for example, due to the deposition of defects at the interface between the dielectric material of the dielectric layer and one of the electrodes. Defects that deposition at the interface influence the barrier height. Knowing the initial barrier height and the parameters that describe the space-charge-limited current, a time-dependent leakage current profile can be mapped. It is also possible to describe the leakage current profile before and after an initial dielectric breakdown. Due to the relationship between leakage current degradation and performance degradation, it is also possible to describe the performance loss.

[0019] In one aspect, the leakage current is monitored. In another aspect, the performance of the semiconductor component is monitored.

[0020] It may be possible to evaluate a leakage current level in a monitoring system and define an initial limit value after which a warning is generated and the remaining service life still has a desired residual value.

[0021] It may be possible to determine a point in time during monitoring when a first breakthrough, first peak, has occurred and to issue a warning including an estimate of a remaining lifetime, which results from the development of the leakage current level or the performance loss, particularly depending on defined product-specific limits.

[0022] Because the load profile plays a significant role even after the first limit value has been reached, several limit values ​​can be defined and active monitoring can continue.

[0023] Figure 1Figure 1 is a schematic representation of a semiconductor device 100. The semiconductor device 100 comprises a first electrode 102, a second electrode 104, and a dielectric layer 106.

[0024] The dielectric layer 106 is arranged between the first electrode 102 and the second electrode 104. In this example, the first electrode 102 and the second electrode 104 are arranged on opposite sides of the dielectric layer 106.

[0025] In the example, the dielectric layer 106 contains defects D1 of a first defect type, defects D2 of a second defect type, and defects D3 of a third defect type. It is also possible that only defects D1 of the first defect type, or only defects of two different defect types, are present. Defects of more than three defect types are also possible. In the example, these defects are arranged in an initial position. This initial position is defined by a manufacturing process of the dielectric layer. The defects are arranged at different distances from the first electrode 102 and the second electrode 104, respectively.

[0026] The defects are charged and, when a potential is applied between the first electrode 102 and the second electrode 104, migrate according to their charge to a specific interface between the dielectric layer 106 and either the first electrode 102 or the second electrode 104. This interface is referred to as the "interface" in the following. This transport of the charged defects is determined by defect properties and a "hopping mechanism".

[0027] In this context, the hopping mechanism means that a displacement, i.e., a hopping, of the defects takes place within an electric field E acting on the dielectric layer 106 of thickness d, determined by the respective applied voltage U. The defects of a defect type i move along localized defect states with a mean effective spacing. AI This leads to a speed vithe movement of defects of defect type i. The speed vi is described using the well-known approach of Variable Range Hopping: v i = C 0 , i a i e − E A , 0 , i k B T sinh N q , i a i E k B T mit E = U d C 0 , i a i = v 0 a i e − 2 a i α

[0028] This represents C 0 ,i ( AI ) a function that describes the influence of a local defect distribution on a velocity characteristic in the dielectric layer 106 vi The defect distribution represents i. This defect distribution is a property of the dielectric layer 106. This property determines the movement of defects of type i in the dielectric layer 106. The parameters decay length α , mean effective distance AI Activation energy E A ,0, i , electric charge N q,i The defects of defect type i are physical properties. k Bis denoted as the Boltzmann constant. T denotes the temperature of the environment of the defect under consideration, in particular the temperature in the dielectric layer 106.

[0029] Up to dielectric failure, a leakage current density can be observed over time. J TED , can be determined using the equation of the thermionic emission diffusion theory according to Crowell and Sze: J TED = qN C v R 1 + v R v D e − Φ B eff k B T e qU k B T − 1

[0030] Here, q represents the unit charge, NC the effective density of states in the conduction band, v R the effective recombination rate, v D the effective diffusion rate, Φ B eff the effective Schottky barrier, k B the Boltzmann constant, T the ambient temperature and U the voltage across the dielectric layer 106.

[0031] Defects i with a positive charge migrate to the electrode with a negative potential and accumulate at the interface. Defects i with a negative charge move to the electrode with a positive potential and accumulate at the interface. This changes the effective Schottky barrier. Φ B eff .

[0032] A barrier height change ΔΦ i , which are generated by defects i, is characterized by their maximum height δ Φ i and a characteristic time constant τ i The characteristic time constant τ i defines a time period during which the barrier height change occurs. Δ Φ i The most significant changes over time t are: Δ Φ i t = δ Φ i 1 − e t τ i

[0033] The maximum height δ Φ i is a function of a number Z i of defects i and depends on a type of interface. With the term ( 1 − e t τ i The time-dependent accumulation of a statistical distribution of defects i in the material is described, in particular the time of accumulation of the respective initial position. In this example, the initial position represents a center point of the distribution. If there are several clusters for defects of type i, one of the clusters can be selected to define the initial position instead of the center point. Preferably, the cluster that has the greatest impact on increasing the service life is selected.

[0034] Various types of defects can be present in the dielectric layer 106. Defects may exist which, due to their charge, effectively compromise the Schottky barrier. Φ B eff increasing barrier height change Δ Φ i This causes the leakage current level to decrease. In this context, these are referred to as "restorative" defects. Defects can exist which, due to their charge, reduce the effective Schottky barrier. Φ B eff reducing barrier height change Δ Φ i This causes the leakage current level to increase. In this context, these are referred to as "degrading" defects.

[0035] When a voltage is applied, the defects are moved. The activation energy and charge can cause certain defects to move much faster.

[0036] In the example, an initial position of defects i is defined by a distance. di the center of gravity of their distribution towards the interface, towards which the defects i tend during the intended operation of the semiconductor component 100.

[0037] The time constant τ iThe time constant can differ for motion under the influence of a first voltage U1 compared to motion under the influence of a second voltage U2. τ i The mobility of defects i under the influence of the electric field E in the dielectric layer 106 and the path to be traveled in the dielectric layer 106 until the respective interface is reached can be defined. During displacement within the dielectric layer 106, defect type i must maintain a distance di the center of gravity of its distribution travels back towards the interface. Along with the speed vi the characteristic time constant results τ i for the deposition process of defects i in τ i = d i v i

[0038] From a measurement of the leakage current over time and evaluation of the time-dependent course using the equations of the thermionic emission diffusion theory according to Crowell and Sze and the time-dependent behavior of the barrier height change, the time constant is determined. τ i Determinable. By means of several measurements of τ i Under different temperatures T and voltages U, the activation energy can E A, 0, i and the electric charge N q,i The defects of defect type i are determined.

[0039] When the first electrical voltage U1 with a first polarity is applied, the charged defects move in a first direction towards the first electrode 102. When the second electrical voltage U2 with a polarity opposite to the first electrical voltage U1 is applied, these charged defects move in a second direction towards the second electrode 102.

[0040] Charged defects that reach the interface at the first electrode 102 or the second electrode 104 accumulate on the interface.

[0041] The semiconductor device 100 can have more than two electrodes. The semiconductor device 100 can have more than one dielectric layer located between each pair of electrodes.

[0042] Dielectric breakdown is possible in such semiconductor devices when operated under an applied electrical voltage. If dielectric breakdown occurs, the interface is partially or completely destroyed. A dielectric breakdown reduces the lifetime of the semiconductor device by 100%. Dielectric breakdown occurs, for example, when the number of charged defects that accumulate at the interface exceeds the voltage-dependent threshold.

[0043] By optimizing process conditions and / or process control during the production of dielectric layer 106, physical properties and production-specific influences on these physical properties can be affected. This can extend the service life.

[0044] Process conditions and / or process control influence, for example, growth, growth conditions, and material composition. These conditions and / or process control can lead to, for example, intentional or unintentional doping and / or contamination during the growth process or in subsequent processes.

[0045] In Figure 2 Figure 1 represents a first state for a first distribution p of defects over a distance x from the second electrode 104. In this example, the first state is an initial state. A distribution of defects D1 of the first defect type is shown in Figure 2. Figure 2and the following figures are labelled with a triangle symbol. A distribution of defects D2 of the second defect type is shown in Figure 2 and the following figures are marked with a star symbol. A distribution of defects D3 of the third defect type is shown in Figure 2 and the following figures are designated with a diamond symbol.

[0046] In Figure 3 A second state is shown for the first distribution p of defects over the distance x from the second electrode 104. In the example, the second state is an operating state following the initial state in the operation of the semiconductor device 100 with the first voltage U1, before a first breakdown, in which defects D2 of the second defect type are already accumulating at the interface of the first electrode 102.

[0047] In Figure 4A third state is shown for the first distribution p of defects over the distance x from the second electrode 104. In this example, the third state is an operating state following the second state in the operation of the semiconductor device 100 with the first voltage U1, before a first breakdown, in which defects D2 of the second defect type and defects D3 of the third defect type are already accumulating at the interface of the first electrode 102. In this situation, it is still possible in this example to extend the lifetime of the semiconductor device by measures that move the defects away from the interface.

[0048] In Figure 5A fourth state is shown for the first distribution p of defects over the distance x from the second electrode 104. In the example, the fourth state is an operating state following the third state in the operation of the semiconductor device 100 with the first voltage U1, before a first breakdown, in which defects D2 of the second defect type and defects D3 of the third defect type are already accumulating at the interface of the first electrode 102, and defects D1 of the first defect type are reaching the interface. In this situation, it is still possible in the example to issue a warning before a performance loss occurs.

[0049] In Figure 6A fifth state is shown for the first distribution p of defects over the distance x from the second electrode 104. In the example, the fifth state is an operating state following the fourth state in the operation of the semiconductor device 100 with the first voltage U1, in which defects D1 of the first defect type, defects D2 of the second defect type, and defects D3 of the third defect type accumulate at the interface of the first electrode 102, and a performance loss occurs.

[0050] In Figure 7 The graph shows a leakage current level I over time t for early detection of remaining service life. In the graph shown... Figure 7 In the example shown, the performance loss from the initial state occurs after approximately 10⁶ seconds of operation with the first voltage U1.

[0051] In Figure 7 The leakage current levels I for the [unclear] are examples of the [unclear] for the [unclear] Figures 2 to 6The described states are marked. The first state is marked with 0, the second state with 1, the third state with 2, the fourth state with 3, and the fifth state with 4.

[0052] For early detection of the remaining service life, an exemplary leakage current level of 700 is provided, upon reaching which, for example, a warning is issued.

[0053] In Figure 8 The graph shows a leakage current level I over time t for the introduction of an electrical load with the second voltage U2 to reverse the defect accumulation. In the graph shown... Figure 8 In the example shown, the performance loss from the initial state occurs after approximately 10⁶ seconds of operation with the first voltage U1.

[0054] In Figure 8 The leakage current levels I for the [unclear] are examples of the [unclear] for the [unclear] Figures 2 to 6The described states are marked. The first state is marked with 0, the second state with 1, the third state with 2, the fourth state with 3, and the fifth state with 4.

[0055] For the electrical load with the second voltage U2 for the reversal of the defect accumulation, an exemplary leakage current level of 800 is provided, at which point this process is started.

[0056] In Figure 9 Figure 1 represents a first state for a second distribution p of defects over a distance x from the second electrode 104. In the example, the first state is an initial state. A distribution of defects D1 of the first defect type is shown in Figure 2. Figure 9 and the following figures are labelled with a triangle symbol. A distribution of defects D2 of the second defect type is shown in Figure 9and the following figures are marked with a star symbol. A distribution of defects D3 of the third defect type is shown in Figure 9 and the following figures are designated with a diamond symbol.

[0057] In Figure 10 A second state is shown for the second distribution p of defects over the distance x from the second electrode 104. In the example, the second state is an operating state following the initial state in the operation of the semiconductor device 100 with the first voltage U1, before a first breakdown, in which defects D2 of the second defect type are already accumulating at the interface of the first electrode 102.

[0058] In Figure 11A third state is depicted for the second distribution p of defects over the distance x from the second electrode 104. In this example, the third state is an operating state following the second state during the operation of the semiconductor device 100 with the first voltage U1, occurring at an early stage of performance loss, in which defects D2 of the second defect type and defects D3 of the third defect type are already accumulating at the interface of the first electrode 102. In this situation, it is possible in this example to trigger an early warning, prompting maintenance of the semiconductor device 100. The threshold for the early warning is chosen, for example, such that the warning is triggered when approximately 10% of the lifetime of the semiconductor device 100 has elapsed from the initial state.The performance loss is not yet noticeable in this example, as each individual breakdown typically only damages a very small area. While this is detectable in the current, it is not noticeable with regard to the function of a product, e.g., an actuator product that includes the semiconductor component 100.

[0059] In Figure 12A fourth state is shown for the second distribution p of defects over the distance x from the second electrode 104. In this example, the fourth state is an operating state following the third state during the operation of the semiconductor device 100 with the first voltage U1, prior to a first breakdown. In this state, defects D2 of the second defect type and defects D3 of the third defect type are already accumulating at the interface of the first electrode 102, and defects D1 of the first defect type are reaching the interface. In this situation, it is possible in this example to trigger a critical warning, which necessitates maintenance of the semiconductor device 100. The threshold for the early warning is chosen, for example, such that the warning is triggered when approximately 50% of the lifetime of the semiconductor device 100 has elapsed from the initial state. In this example, several local breakdowns have already occurred in this state.The leakage current level is then significantly higher and detectable. The total area destroyed by numerous local breakdowns is now noticeable in the percentage range of the product's functionality. Therefore, a warning is issued, for example. In this example, the semiconductor component 100, or the product, is still well on the verge of failure.

[0060] In Figure 13 A fifth state is shown for the second distribution p of defects over the distance x from the second electrode 104. In this example, the fifth state is an operating state following the fourth state during the operation of the semiconductor device 100 with the first voltage U1. In this state, defects D1 of the first defect type, defects D2 of the second defect type, and defects D3 of the third defect type accumulate at the interface of the first electrode 102, resulting in a significant performance loss. In this example, this is the point at which the lifetime of the semiconductor device 100 ends.

[0061] The end of the service life refers to the time at which the semiconductor component 100 can no longer be used as intended. This limit can vary for different applications.

[0062] In Figure 14 The graph shows the course of a leakage current level I over time t. In the graph shown... Figure 14 In the example shown, the performance loss increases from the initial state and reaches the end of the service life after approximately 9 x 10⁴ seconds of operation at the first voltage U1. The end of the service life depends on the material and the conditions. In this example, the end of the service life occurs later than the fourth state. However, a performance loss may still be slight even after reaching the fifth state.

[0063] In Figure 14 The leakage current levels I for the [unclear] are examples of the [unclear] for the [unclear] Figures 9 to 13The described states are marked. The first state is marked with 0, the second state with 1, the third state with 2, the fourth state with 3, and the fifth state with 4.

[0064] For a critical warning, an exemplary leakage current level of 1400 is provided, upon reaching which, for example, the critical warning is issued.

[0065] In Figure 15 A device 200 for manufacturing the semiconductor component 100 according to the first embodiment is shown schematically.

[0066] The device 200 comprises a control and / or monitoring unit 202 configured to perform steps in a procedure described below for monitoring the semiconductor device 100. The device 200 is at least temporarily connectable to the first electrode 102 of the semiconductor device 100 via a first conductor 204. The device 200 is at least temporarily connectable to the second electrode 104 of the semiconductor device 100 via a second conductor 206.

[0067] In this example, the control and / or regulating device 202 is configured to output a first voltage U1 with a first polarity. The first voltage U1 is present between the first electrode 102 and the second electrode 104 when these are connected to the device 200 via their respective conductors and the control and / or regulating device 202 outputs the first voltage U1.

[0068] The control and / or regulating device 202 is optionally configured to output a second voltage U2 with a polarity opposite to that of the first. In this example, the second voltage U2 is present between the first electrode 102 and the second electrode 104 when these are connected to the device 200 via their respective conductors and the control and / or regulating device 202 outputs the second voltage U2.

[0069] The control device can be configured to output an alternating voltage or a substantially constant value for the first voltage U1 and / or the second voltage U2. This value is, for example, in a range of 1 volt to 80 volts and is preferably 1 volt, 2 volts, 5 volts, 10 volts, 20 volts, 40 volts, or 80 volts.

[0070] Preferably, the dielectric layer 106 is designed as a polycrystalline oxide high-k dielectric.

[0071] The dielectric layer 106 is specifically designed as a PZT layer. PZT here denotes Pb(Zr x Ti 1- ×)O 3 .

[0072] The dielectric layer 106 is specifically designed as a KNN layer. KNN here denotes (K x Na 1-x )NbO 3 .

[0073] The dielectric layer 106 can in particular be formed as an HfO 2 -, HfZrO 2 -, ZrO 2 -, BaTiO 3 -, SrTiO 3 - or (Ba x Sr 1-x )TiO 3 - layer.

[0074] Preferably, the dielectric layer 106 is doped. For example, the dielectric layer 106 is a PZT layer doped with nickel: Pb(Zr x Ti 1-x )O 3 Ni y .

[0075] Both the PZT layer and the KNN layer can have dopants other than nickel, for example Nb, La, Mn, Mg.

[0076] Preferably, the dielectric layer is formed as a sputtered PZT layer. In this process, the target material is deposited onto a substrate in a plasma. PZT is used, for example, as the target material. Preferably, the sputtered PZT layer has a deposition temperature of less than 500°C.

[0077] The preferred thickness of the dielectric layer is in the range of 500 nm to 4 µm. This is a practical range for actuators. Preferably, the layer thickness is 1 µm or 4 µm. Larger layer thicknesses are also possible. The described procedure can be carried out for all layer thicknesses.

[0078] Reduced layer thicknesses are also possible. A dielectric layer thickness of less than 500 nm is preferred, for example, for applications outside of actuators. One example of this is the application of high-k dielectrics, e.g., as memory. The memory can be a resistive random access memory (ReRAM) or a ferroelectric random access memory (FeRAM). For these other applications, thicknesses of 15 nm to 200 nm are suitable, depending on the specific application.

[0079] Additionally, very thin high-k dielectrics with layers, e.g., as gate oxides, can be used in a wide variety of applications. For example, HfO2 or SiO2 with a layer thickness of less than or equal to 50 nm can be used as dielectric layer 106.

[0080] The control device can be configured to regulate a setpoint for the first voltage U1. This setpoint is, for example, in a range of 1 volt to 80 volts and is preferably 1 volt, 2 volts, 5 volts, 10 volts, 20 volts, 40 volts, or 80 volts.

[0081] The control device can be configured to regulate a setpoint for the second voltage U2. This setpoint is, for example, in a range of 1 volt to 80 volts and is preferably 1 volt, 2 volts, 5 volts, 10 volts, 20 volts, 40 volts, or 80 volts.

[0082] In this case, a voltage measuring device can be provided, which is configured to detect the voltage between the first electrode 102 and the second electrode 104. The control device is configured in this case to reduce a control deviation, which is determined as a function of the difference between this detected voltage and the setpoint.

[0083] In the example, the device 200 optionally includes a measuring device 208. In the example, the measuring device 208 is configured to measure the current flowing through the first electrode 102 and the second electrode 104.

[0084] In this example, the measuring device 208 is arranged in the first conductor 204. A corresponding arrangement at another location is also possible.

[0085] In the example, the measuring device 208 is connected to the control and / or regulating device 202 via a signal line 210 for the transmission of the measured current.

[0086] The leakage current is strongly temperature-dependent. The measuring device 208 may include a device for determining the current temperature T. Temperature measurement is performed on the semiconductor device 100 or in its immediate vicinity. The limit value G at temperature T is determined in this respect as a function of a reference limit value G0 at a reference temperature T0 according to the following exponential relationship: G T , T 0 = G 0 ∗ e − k k B T e − k k B T 0 k is a constant here. This constant describes the effective barrier height at this point in time.

[0087] In this example, the device 200 comprises a voltage and / or current source 212 for outputting the voltages. In this example, the device 200 comprises a computing unit 214, in particular a microprocessor, configured to execute the methods described below. In this example, the device 200 comprises a memory 216 configured to store limit values ​​or a characteristic curve from which limit values ​​can be determined. In this example, the device 200 comprises an output unit 218 configured to output a value 220.

[0088] This device for monitoring the semiconductor device 100 and the semiconductor device 100 itself can include a memory, actuator, MEMS, micromirror, printhead, or loudspeaker. It can be provided for lifetime monitoring of the memory, actuator, MEMS, micromirror, printhead, or loudspeaker, in which the semiconductor device 100 is monitored.

[0089] The procedure for monitoring the semiconductor device 100 is described below using the following example: Figure 16 described.

[0090] In step 1600, the semiconductor device 100 is operated in a first operating mode in which the first voltage U1 is applied.

[0091] In step 1602, a leakage current I is detected during the operation of the semiconductor device 100, which flows through the first electrode 102 and the second electrode 104 of the semiconductor device 100.

[0092] In step 1604, the leakage current I is compared to a first limit value, e.g., one of the leakage current levels 700, 800, or 1400. Depending on the result of this comparison, the output 220 is determined in step 1604. The first limit value can be selected from a range of limits that the leakage current I exceeds or falls below.

[0093] It may be possible to take into account a voltage and / or temperature dependence of the leakage current I. For example, the limit value is determined depending on the reference limit value G0 and the reference temperature T0, in particular using the exponential relationship described above.

[0094] In step 1604, a time point can be determined in one aspect at which an extremal point, in particular a maximum, of the leakage current I occurs. In this aspect, the output 220 can be determined as a function of the time point.

[0095] Output 220 comprises one state of the semiconductor device 100.

[0096] According to the invention, the residual lifetime of the semiconductor device 100 is determined depending on the time at which the extremum, in particular the maximum, of the leakage current I occurs. In this case, the output 220 can include this residual lifetime.

[0097] For example, a text is defined that includes the condition of the semiconductor device 100 and information about its remaining lifetime, e.g., the residual value. In this example, the residual value is the duration for which the semiconductor device 100 can still be operated with the first voltage U1, starting from the time the leakage current I is detected, until the time at which the lifetime of the semiconductor device 100 ends. This duration is calculated, for example, starting from the time the leakage current I is detected, using the previously described relationship for defect movement under the influence of the first voltage U1, until the relevant end-of-life time for the respective application.

[0098] In one aspect, a control signal is determined that specifies an operating mode for the semiconductor device 100.

[0099] In step 1606, the output 220 is displayed.

[0100] For example, the text is sent in a message or output at a human-machine interface.

[0101] For example, the control signal is output to control the semiconductor device 100 or a device comprising the semiconductor device 100 to operate in the operating mode.

[0102] Step 1600 is then executed, specifically using the specified operating mode. The procedure may be terminated for one of the reasons described below.

[0103] It may be provided that the leakage current I is detected in the first operating mode of the semiconductor device 100 and that the operation of the semiconductor device 100 in step 1602 is continued in the first operating mode if the leakage current I exceeds the first limit value or if the remaining lifetime is greater than a first threshold value.

[0104] It can be provided that, in this case, a second limit value is subsequently determined from the multitude of limit values, and the operation of the semiconductor device 100 either continues in a second operating mode or the operation is terminated if the leakage current I exceeds the second limit value. Thus, various measures are taken as the leakage current level increases.

[0105] Alternatively, the operation of the semiconductor device 100 can be either continued in a second operating mode or terminated. Operation can continue in the second operating mode if the remaining lifetime is less than a second threshold. Operation can be terminated if the remaining lifetime is greater than the second threshold. Various measures can be taken as the remaining lifetime decreases.

Claims

1. Method for monitoring a semiconductor component (100), wherein a leakage current (I), which flows through a first electrode (102) and a second electrode (104) of the semiconductor component (100), is detected (1602) during operation of the semiconductor component (100), wherein a time at which an extreme point, in particular a maximum, of the leakage current (I) occurs is determined (1604) and an output (220) is determined based on the time, wherein the output includes a state of the semiconductor component (100), and the output (220) is output (1606) and wherein optionally the leakage current (I) is compared with a first limit value (700, 800, 1400) for the leakage current (I) in a comparison and an output (220) is determined (1604) based on a result of the comparison, characterized in that a residual value of a lifetime of the semiconductor component (100) is determined (1604) based on the time at which the extreme point, in particular the maximum, of the leakage current (I) occurs, wherein the output (220) includes the residual value.

2. Method according to Claim 1, characterized in that the leakage current (I) is detected (1602) during operation of the semiconductor component (100) in a first operating mode, wherein the operation of the semiconductor component (100) is continued (1602) in the first operating mode when the leakage current (I) exceeds the first limit value (700, 800) or when the residual value of the lifetime is greater than a first threshold value.

3. Method according to Claim 2, characterized in that a second limit value (1400) is determined from the plurality of limit values, wherein the operation of the semiconductor component (100) is continued (1602) in a second operating mode or the operation is terminated when the leakage current (I) exceeds the second limit value (1400) or in that the operation of the semiconductor component (100) is continued (1602) in a second operating mode or the operation is terminated when the residual value of the lifetime is greater than a second threshold value.

4. Method according to one of the preceding claims, characterized in that a reference limit value and a reference temperature are specified, a current temperature of the semiconductor component (100) or of surroundings of the semiconductor component (100) is determined, a factor is determined based on the current temperature and the reference temperature, the reference limit value is scaled by the factor and the first limit value and / or the second limit value is determined based on the reference limit value scaled by the factor.

5. Method according to one of the preceding claims, characterized in that a text, which includes the state of the semiconductor component (100) and an indication of a still remaining lifetime, is determined, wherein the text is sent in a message or is output to a human-machine interface (1606).

6. Method according to one of the preceding claims, characterized in that a control signal, which specifies an operating mode for the semiconductor component (100), is determined (1604), wherein the control signal is output (1606) in order to actuate the semiconductor component (100) or a unit comprising the semiconductor component for operation in the operating mode.

7. Device (200) for monitoring a semiconductor component (100), wherein the device (200) comprises a measurement unit (208), a computation unit (214) and an output unit (218), wherein the measurement unit (208) is designed to detect a leakage current (I), which flows through a first electrode (102) and a second electrode (104) of the semiconductor component (100), during operation of the semiconductor component (100), wherein the computation unit (214) is designed to compare the leakage current (I) with a first limit value (700, 800, 1400) for the leakage current (I) in a comparison and to determine an output (220) based on a result of the comparison and / or wherein the computation unit (214) is designed to determine a time at which an extreme point, in particular a maximum, of the leakage current (I) occurs and to determine an output (220) based on the time, wherein the output includes a state of the semiconductor component (100), and wherein the output unit (218) is designed to output the output (220), characterized in that the computation unit is designed to determine a residual value of a lifetime of the semiconductor component (100) based on the time at which the extreme point, in particular the maximum, of the leakage current (I) occurs, wherein the output (220) includes the residual value.

8. MEMS, memory, actuator, micromirror, printhead or loudspeaker comprising the device (200) for monitoring a semiconductor component (100) according to Claim 7 and the semiconductor component (100).