Simulation system, simulation method, and simulation program

Abstract

This simulation system acquires, for each of a plurality of energy values constituting a cosmic-ray energy distribution, a first profile that indicates the number of events, which is the count of absorptions of cosmic rays having that energy value by a transistor per unit time, converts each of the plurality of energy values into a charge amount to convert the first profile into a second profile that indicates the number of events for a respective charge amount, performs a circuit simulation in which a forward current from the transistor that is based on the specification of the transistor and an error current that is determined on the basis of the second profile are repeatedly compared a plurality of times, and identifies, as a soft error, a case in which the error current is equal to or greater than the forward current in the circuit simulation.

Description

Simulation system, simulation method, and simulation program

[0001] One aspect of this disclosure relates to a simulation system, a simulation method, and a simulation program.

[0002] A method for estimating soft errors caused by cosmic rays using simulations is known. For example, Patent Document 1 describes a simulation device comprising: a discrete event simulation unit that performs discrete simulations of the components of a defined configuration model based on attribute information, which is component information of the components of the defined configuration model, and connection information, which indicates the connection relationships between the components of the defined configuration model; and a soft error rate calculation processing unit that calculates the soft error rate of the defined configuration model based on the simulation results and soft error rate data in the attribute information.

[0003] Patent Document 2 describes a method for calculating the occurrence rate of soft errors in semiconductor integrated circuits. This method includes a distance calculation step of calculating the distance along a well where a parasitic bipolar effect can occur as a limit distance, starting from a diffusion layer where a funneling phenomenon has occurred, and a step of determining that a soft error has occurred in a cell containing a transistor if an off-state transistor exists within the limit distance along the well from the diffusion layer where a funneling phenomenon has occurred due to the passage of charged particles.

[0004] Patent Document 3 describes a method for analyzing errors caused by cosmic ray neutrons. This method, when simulating failures of semiconductor devices caused by cosmic ray neutrons, includes the steps of: defining a unit cell of a device as a collection of components; analyzing the orbits of secondary ions produced as a result of nuclear reactions between cosmic ray neutrons and atomic nuclei constituting the device on a component-by-component basis; and determining the amount of charge collected in the device's diffusion layer, the collection cell, the location of the diffusion layer, and whether or not a failure occurred for a number of nuclear reaction events occurring at specific locations within the device.

[0005] Japanese Patent Application Laid-open No. 2011-257898, Japanese Patent Application Laid-open No. 2010-205048, Japanese Patent Application Laid-open No. 2004-138529

[0006] A technique for obtaining soft errors with higher accuracy is desired.

[0007] A simulation system according to one aspect of the present disclosure includes at least one processor. The at least one processor acquires, for each of a plurality of energy values constituting the energy distribution of cosmic rays, a first profile indicating the number of events, which is the number of times per unit time that cosmic rays having the energy value are absorbed by the transistor, converts each of the plurality of energy values into a charge amount, converts the first profile into a second profile indicating the number of events for each of a plurality of charge amounts, executes a circuit simulation that repeats multiple comparisons between a forward current from the transistor based on transistor specifications and an error current determined based on the second profile, and identifies a case where the error current is equal to or greater than the forward current as a soft error in the circuit simulation.

[0008] In this aspect, the first profile indicating the relationship between the energy distribution of cosmic rays and the number of events is converted into the second profile indicating the relationship between the distribution of charge amounts affecting the transistor and the number of events. Then, soft errors are identified based on the results of circuit simulation based on transistor specifications and the second profile. By identifying soft errors in consideration of the energy distribution of cosmic rays in this way, soft errors can be obtained with higher accuracy.

[0009] According to one aspect of the present disclosure, soft errors can be obtained with higher accuracy.

[0010] It is a diagram showing an example of the functional configuration of the simulation system. It is a flowchart showing an example of the operation of the simulation system. It is a diagram showing an example of conversion from a first profile to a second profile. It is a diagram showing a structural model according to an embodiment. It is a diagram showing a method for setting the arrangement of particles in a sealing material in an embodiment.

[0011] Hereinafter, various examples of the present disclosure will be described in detail with reference to the accompanying drawings. In the description of the drawings, identical or equivalent elements are denoted by the same reference numerals, and redundant descriptions are omitted.

[0012] [System Configuration] The simulation system according to the present disclosure is a computer system that estimates soft errors occurring in electronic components caused by cosmic rays using simulation. Cosmic rays refer to radiation traveling through outer space. A soft error refers to a transient malfunction occurring in a semiconductor chip caused by radiation entering the semiconductor chip. Unlike hard errors caused by physical damage to a semiconductor chip, soft errors can be resolved by measures such as restarting the device or rewriting data.

[0013] The simulation system executes the aforementioned simulation for an electronic component including a sealing material containing a plurality of particles having a cosmic ray absorbing capability, and one or more transistors (elements) protected by the sealing material. The one or more transistors are assumed as components of a semiconductor chip. In one example, the simulation system is used to search for a configuration of a sealing material that can reduce soft errors. Hereinafter, particles having a cosmic ray absorbing capability are also simply referred to as "particles". These particles are also referred to as neutron absorbers. In one example, the particles in the sealing material are gadolinium oxide (Gd ２ O ３ ), samarium oxide (Sm ２ O ３ ), and one type of metal oxide selected from boron nitride (BN).

[0014] The simulation system is composed of one or more computers. When a plurality of computers are used, these computers are connected via a communication network such as the Internet or an intranet, thereby logically constructing a single simulation system.

[0015] A computer comprising a simulation system generally includes a processor, memory, a communication interface, input devices, and output devices as hardware components. Examples of processors include CPUs and GPUs. Memory can consist of flash memory, hard disks, etc. The communication interface can consist of a network card or wireless communication module. Examples of input devices include keyboards, pointing devices, touch panels, microphones, sensors, and cameras. Examples of output devices include monitors, touch panels, head-mounted displays (HMDs), and speakers. Each functional module of the simulation system is realized by the processor executing programs stored in memory. Data or databases necessary for processing each functional module of the simulation system may also be stored in memory.

[0016] A simulation program for enabling a computer to function as a simulation system includes program code for implementing each functional module of the simulation system. This simulation program may be provided on a non-temporary recording medium such as a CD-ROM, DVD-ROM, or semiconductor memory. Alternatively, the simulation program may be provided via a communication network as a data signal superimposed on a carrier wave. The provided simulation program is then recorded, for example, on a storage device.

[0017] Referring to Figure 1, the configuration of an example simulation system 10 will be described. Figure 1 is a diagram showing the functional configuration of the simulation system 10.

[0018] The simulation system 10 includes a processor 20. In one example, the processor 20 functions as a model generation unit 11, a cosmic ray simulation unit 12, a conversion unit 13, and a circuit simulation unit 14.

[0019] The model generation unit 11 is a functional module that generates a structural model, which is data representing an electronic component comprising a sealing material and a transistor.

[0020] The cosmic ray simulation unit 12 is a functional module that performs a cosmic ray simulation by irradiating its structural model with cosmic rays. As a result of the simulation, the cosmic ray simulation unit 12 generates a first profile that shows the number of times per unit time that cosmic rays are absorbed by the transistor. The first profile is data that shows the number of events per unit time for each of the multiple energy values ​​that constitute the energy distribution of cosmic rays, which is the number of times per unit time that cosmic rays with that energy value are absorbed by the transistor. In other words, the first profile shows the relationship between the energy distribution of cosmic rays and the number of events.

[0021] The conversion unit 13 is a functional module that converts each of the multiple energy values ​​into charge quantities and converts the first profile into a second profile. The second profile is data showing the number of events for each of the multiple charge quantities. In other words, the second profile shows the relationship between the distribution of charge quantities and the number of events.

[0022] The circuit simulation unit 14 is a functional module that performs circuit simulations by repeatedly comparing the forward current from a transistor based on the transistor's specifications with the error current determined based on a second profile. The circuit simulation unit 14 identifies a soft error when the error current is greater than or equal to the forward current in the circuit simulation. The circuit simulation unit 14 can also perform a process to calculate a predetermined index based on the identified soft error.

[0023] [System Operation] The operation of the simulation system 10 will be described as an example of the simulation method relating to this disclosure, with reference to Figure 2. Figure 2 is a flowchart showing an example of its operation as the processing flow S1.

[0024] In step S11, the model generation unit 11 generates a structural model representing the electronic component. For example, the model generation unit 11 provides the user with a predetermined user interface, accepts user input through that user interface, and generates a structural model based on that user input.

[0025] As described above, the structural model represents an electronic component including a sealing material containing a plurality of particles having an ability to absorb cosmic rays, and one or more transistors protected by the sealing material. The structural model includes a transistor layer representing one or more transistors, and a sealing material layer laminated above the transistor layer and representing the sealing material. As described above, the sealing material layer includes Gd ２ O ３ , Sm ２ O ３ , and BN, and can represent a sealing material containing particles composed of one of the foregoing. The structural model may further include a substrate layer representing a substrate, an insulating layer representing an insulator, and a metal layer representing a wiring. In one example, the model generating unit 11 generates a transistor layer that simulates a configuration in which a plurality of transistors are two-dimensionally arranged at predetermined intervals along a horizontal direction. In the present disclosure, a region that represents a transistor and constitutes the transistor layer is referred to as a "sensitive region". In one example, the model generating unit 11 generates the transistor layer such that each sensitive region straddles the substrate layer and the insulating layer.

[0026] In one example, the model generating unit 11 accepts a user input that specifies the concentration of the plurality of particles in the sealing material. Then, the model generating unit 11 generates the sealing material layer of the structural model such that the arrangement of the plurality of particles in the sealing material satisfies the specified concentration. The unit of the particle concentration may be mass percent or volume percent.

[0027] In step S12, the cosmic ray simulation unit 12 executes a cosmic ray simulation using the structural model. Cosmic ray simulation refers to a process that virtually implements, on a computer, a scenario where an electronic component is exposed to cosmic rays in outer space such as on an artificial satellite or a space station. As the cosmic ray simulation, the cosmic ray simulation unit 12 virtually irradiates the structural model with a plurality of cosmic rays such that the plurality of cosmic rays pass through the sealing material layer and the transistor layer in this order. For example, the cosmic ray simulation unit 12 simulates a scenario where a plurality of cosmic rays are incident on the upper surface of the sealing material layer from above the structural model. In one example, the cosmic ray simulation unit 12 １ H, ４ He,７ Li, ９ Be, １１ B, １２ C, １４ N, １６ O, ２０ Ne, ２４ Mg, ２８ Si, and ５６ Multiple cosmic rays, each originating from a different isotope of Fe, are virtually irradiated onto the structural model. That is, each cosmic ray is radiation originating from one of the 12 isotopes. The cosmic ray simulation unit 12 randomly determines the energy value of each cosmic ray within a predetermined energy range and virtually irradiates the structural model with the cosmic ray having the determined energy value. In the cosmic ray simulation, each cosmic ray may be absorbed by particles in the encapsulating layer and not reach the transistor layer, or it may pass through the sensitive region (transistor) without being absorbed in the encapsulating layer, or it may not be absorbed in the encapsulating layer and may proceed without passing through the sensitive region.

[0028] In step S13, the cosmic ray simulation unit 12 generates a first profile showing the relationship between the energy distribution of cosmic rays and the number of events. The cosmic ray simulation unit 12 determines that a cosmic ray has been absorbed by a transistor when a virtually irradiated cosmic ray passes through a sensitive region (transistor) in the structural model. The cosmic ray simulation unit 12 generates the first profile based on data showing cosmic rays that have passed through the sensitive region in the cosmic ray simulation. In one example, the cosmic ray simulation unit 12 sets the unit time for obtaining the number of events to 1 second.

[0029] In step S14, the conversion unit 13 converts the first profile into a second profile that shows the relationship between the distribution of charge and the number of events. This process can be described as a conversion from the energy distribution of cosmic rays to the distribution of charge. The conversion unit 13 obtains the first profile generated by the cosmic ray simulation unit 12. The conversion unit 13 converts each of the multiple energy values ​​shown in the first profile into charge, thereby converting the first profile into the second profile. The conversion unit 13 performs the following conversion for each energy value. That is, the conversion unit 13 divides the energy value of the cosmic ray by the energy value required to generate an e-h pair. The energy value required to generate an e-h pair is determined by the transistor material. The conversion unit 13 multiplies the value obtained by the division by the elementary charge to obtain the charge. If (conversion constant) = (elementary charge) / (energy value required to generate an e-h pair), the conversion unit 13 multiplies the energy value of the cosmic ray by this conversion constant to obtain the charge.

[0030] Figure 3 shows an example of conversion from the first profile to the second profile. Figure 3 shows the first profile 31 and the second profile 32 as graphs. In the graph for the first profile 31, the vertical axis shows the number of events per second, and the horizontal axis shows the energy value of the cosmic rays. In the graph for the second profile 32, the vertical axis shows the number of events per second, and the horizontal axis shows the amount of charge. In other words, in the example of Figure 3, the unit time is 1 second.

[0031] Returning to Figure 2, in step S15, the circuit simulation unit 14 performs a circuit simulation based on the transistor specifications and the second profile. Circuit simulation refers to the process of virtually realizing the operation of electronic components in outer space, such as artificial satellites and space stations, on a computer. The circuit simulation unit 14 virtually operates a predetermined circuit having one or more transistors. In addition to the transistor specifications and the second profile, the circuit simulation unit 14 may also operate the predetermined circuit based on the circuit specifications and at least one of the semiconductor properties.

[0032] In the circuit simulation, the circuit simulation unit 14 simultaneously generates a forward current from the transistor based on its specifications and an error current determined based on a second profile at predetermined intervals, and compares these forward currents and error currents. In other words, the circuit simulation unit 14 performs a circuit simulation that repeatedly compares the forward current and error current multiple times.

[0033] Forward current refers to the current generated according to the specifications of the transistor when the circuit is operated. The circuit simulation unit 14 generates the same forward current at predetermined intervals.

[0034] On the other hand, the error current determined based on the second profile may differ each time. The individual charge amounts shown by the second profile are values ​​for the entire transistor layer. The circuit simulation unit 14 divides the number of events by the number of transistors to obtain the number of events per transistor. As a result, the relationship between the charge amount and the number of events for one transistor, i.e., the second profile for one transistor, is obtained. Hereinafter, the second profile for one transistor will be conveniently referred to as the "unit second profile". The circuit simulation unit 14 determines the error current using the unit second profile at predetermined intervals. The circuit simulation unit 14 randomly selects a charge amount from the unit second profile, divides the selected charge amount by a predetermined minute time, and determines the minute current at that minute time as the error current.

[0035] In step S16, the circuit simulation unit 14 identifies soft errors in the circuit simulation. The circuit simulation unit 14 identifies a soft error when the error current is greater than or equal to the forward current, and identifies normal operation when the error current is less than the forward current. The circuit simulation unit 14 repeats this process to obtain the number of soft errors. The number of soft errors obtained through this process is the number per unit time, for example, the number of soft errors that occurred per second.

[0036] In step S17, the circuit simulation unit 14 generates simulation results based on the identified soft errors. For example, the circuit simulation unit 14 calculates the number of soft errors that occurred per second as 10 ９ The soft error rate can be calculated by converting it to the number of errors per unit of time, and simulation results including this soft error rate can be generated. If the number of soft errors that occur per second is n, then the soft error rate (FIT / Mbit) is n × 60 × 60 × 10 ９ This is obtained. Furthermore, the circuit simulation unit 14 calculates the square root of the reciprocal of the number of cosmic rays that contributed to the soft error as the relative standard error of the soft error occurrence rate. As another example, the circuit simulation unit 14 may generate the progression of the cumulative number of soft errors over time as part of the simulation results.

[0037] In step S18, the circuit simulation unit 14 outputs the simulation results. The circuit simulation unit 14 may display the simulation results on a display device, store the simulation results in a storage device, or transmit the simulation results to another computer such as a user terminal.

[0038] As a variation of steps S16 to S18, the circuit simulation unit 14 may output simulation results related to soft errors in real time in accordance with the progress of the circuit simulation.

[0039] [Variations] The technology relating to this disclosure has been described in detail above based on various examples. However, this disclosure is not limited to the above examples. Various modifications are possible to the technology relating to this disclosure without departing from its essence.

[0040] The simulation system may acquire a first profile generated by another computer system and convert that first profile into a second profile. In other words, the simulation system does not need to have functional modules corresponding to the model generation unit 11 and the cosmic ray simulation unit 12 described above.

[0041] The processing steps for a method executed by at least one processor are not limited to the examples above. For example, some of the steps described above may be omitted, or each step may be performed in a different order. Also, any two or more of the steps described above may be combined, or some of the steps may be modified or deleted. Alternatively, other steps may be performed in addition to each of the steps described above.

[0042] In comparing the relative magnitudes of two numerical values ​​in this disclosure, either of the two criteria, "greater than or equal to" and "greater than," may be used, or either of the two criteria, "less than or equal to" and "less than," may be used.

[0043] In this disclosure, the expression "at least one processor executes a first process, a second process, ... and the nth process," or a corresponding expression, refers to a concept that includes cases where the entity executing the n processes from the first process to the nth process, i.e., the processor, changes along the way. In other words, this expression refers to a concept that includes both cases where all n processes are executed by the same processor and cases where the processor changes at an arbitrary rate for the n processes.

[0044] [Note] As can be seen from the various examples above, this disclosure includes the following embodiments: (Note 1) A simulation system comprising at least one processor, the at least one processor acquires a first profile indicating the number of events per unit time, which is the number of times a cosmic ray having a certain energy value is absorbed by a transistor, for each of a plurality of energy values ​​constituting the energy distribution of cosmic rays, converts each of the plurality of energy values ​​into a charge quantity, converts the first profile into a second profile indicating the number of events for each of the plurality of charge quantities, performs a circuit simulation which involves repeatedly comparing a forward current from the transistor based on the specifications of the transistor with an error current determined based on the second profile, and identifies a soft error when the error current is greater than or equal to the forward current in the circuit simulation. (Note 2) The simulation system according to Note 1, wherein at least one processor generates a structural model representing an electronic component comprising a encapsulant containing a plurality of particles having the ability to absorb cosmic rays and a transistor protected by the encapsulant, wherein a encapsulant layer representing the encapsulant is stacked on top of a transistor layer representing the transistor; performs a cosmic ray simulation in which the plurality of cosmic rays are virtually irradiated onto the structural model with the plurality of cosmic rays passing through the encapsulant layer and the transistor layer in that order; generates the first profile based on data showing the cosmic rays absorbed by the transistor in the structural model in the cosmic ray simulation; and obtains the generated first profile. (Note 3) The simulation system according to Note 2, wherein at least one processor receives a user input specifying the concentration of the plurality of particles in the encapsulant, and generates the encapsulant layer of the structural model such that the arrangement of the plurality of particles in the encapsulant satisfies the concentration.(Note 4) The simulation system according to Note 2 or 3, wherein at least one processor generates the structural model to represent the plurality of particles composed of one of gadolinium oxide, samarium oxide, and boron nitride. (Note 5) The at least one processor, １ H, ４ He, ７ Li, ９ Be, １１ B, １２ C, １４ N, １６ O, ２０ Ne, ２４ Mg, ２８ Si, and ５６A simulation system according to any one of Appendix 2 to 4, wherein the plurality of cosmic rays, each of which has an atomic nucleus of Fe as a source, are virtually irradiated onto the structural model. (Appendix 6) A simulation method performed by a simulation system comprising at least one processor, comprising: a step of obtaining a first profile indicating the number of events per unit time, which is the number of times the cosmic ray having the energy value is absorbed by the transistor, for each of a plurality of energy values ​​constituting the energy distribution of cosmic rays; a step of converting each of the plurality of energy values ​​into a charge quantity, and converting the first profile into a second profile indicating the number of events for each of the plurality of charge quantities; a step of performing a circuit simulation in which a comparison of the forward current from the transistor based on the specifications of the transistor and the error current determined based on the second profile is repeated a plurality of times; and a step of identifying as a soft error when the error current is greater than or equal to the forward current in the circuit simulation. (Note 7) A simulation program that causes a computer to perform the following steps: obtaining a first profile showing the number of events per unit time, which is the number of times a cosmic ray having a certain energy value is absorbed by a transistor, for each of a plurality of energy values ​​that constitute the energy distribution of cosmic rays; converting each of the plurality of energy values ​​into a charge quantity, and converting the first profile into a second profile showing the number of events for each of the plurality of charge quantities; performing a circuit simulation that repeatedly compares the forward current from the transistor based on the specifications of the transistor with the error current determined based on the second profile; and identifying a soft error in the circuit simulation when the error current is greater than or equal to the forward current.

[0045] According to appendices 1, 6, and 7, the first profile, which shows the relationship between the energy distribution of cosmic rays and the number of events, is converted into a second profile, which shows the relationship between the distribution of charge affecting the transistor and the number of events. Then, the soft error is identified based on the transistor specifications and the results of circuit simulations based on the second profile. By identifying the soft error in this way, taking into account the energy distribution of cosmic rays, the soft error can be determined with greater accuracy.

[0046] According to Appendix 2, electronic components comprising encapsulating material and transistors are represented by a structural model. Then, a cosmic ray simulation is performed, virtually irradiating the structural model with cosmic rays, to generate a first profile. This series of processes allows for more accurate determination of soft errors in electronic components with encapsulating material. It also makes it possible to evaluate the performance of encapsulating material in absorbing cosmic rays.

[0047] According to Appendix 3, the encapsulant layer of the structural model is generated to satisfy the particle concentration specified by the user input. This configuration makes it possible to obtain information on soft errors for various concentrations. As a result, a preferred encapsulant configuration can be explored.

[0048] According to Appendix 4, the soft error can be determined when the particles in the encapsulant are composed of gadolinium oxide, samarium oxide, or boron nitride, which are substances expected to react readily to cosmic rays.

[0049] According to Appendix 5, representative cosmic rays are considered in the cosmic ray simulation, allowing for a more accurate determination of soft errors.

[0050] The following describes specific examples, but this disclosure is not limited to them.

[0051] The simulation system related to this disclosure is implemented on a computer, Gd ２ O ３ Sm ２ O ３For each of the above, and BN, the effect of controlling or reducing the impact of cosmic rays on electronic components was verified through simulation. The series of steps to obtain the simulation results were as follows.

[0052] (Structural Model) Figure 4 is a diagram showing a structural model according to an embodiment. In the embodiment, a structural model 100 was created in which a substrate layer 110, an insulating layer 120, a metal layer 130, and a sealing material layer 140 are arranged in this order from bottom to top. The substrate layer 110 was set assuming a silicon substrate, and silicon dioxide (SiO ２ An insulating layer 120 was set assuming ) and a metal layer 130 was set assuming copper (Cu) wiring. For the sealing material layer 140, Gd was added to the bisphenol A type epoxy resin. ２ O ３ Sm ２ O ３ We envisioned a encapsulating material in which one type of particle (neutron absorber) selected from BN is scattered.

[0053] The overall dimensions of the structural model 100 were set to width (W), depth (D), and height (H) of 1000 μm, 1000 μm, and 1503.35 μm, respectively. The heights of the substrate layer 110, insulating layer 120, metal layer 130, and sealing layer 140 were set to 489.75 μm, 0.35 μm, 3 μm, and 1000 μm, respectively.

[0054] A transistor layer 150 was set in the region spanning the substrate layer 110 and the insulating layer 120. Silicon was assumed to be the material for each transistor. A sensitive region 151 representing a single transistor was represented by a three-dimensional object that exhibits an inverted T-shape when viewed from the side. As shown in the enlarged view in Figure 4, this inverted T-shaped three-dimensional object is formed by placing an upper rectangular parallelepiped, smaller than the lower rectangular parallelepiped, in the center of the upper surface of the lower rectangular parallelepiped. The dimensions of the lower rectangular parallelepiped were set to 0.2625 μm, 0.147 μm, and 0.5 μm in width, depth, and height, respectively. The dimensions of the upper rectangular parallelepiped were set to 0.175 μm, 0.098 μm, and 0.25 μm in width, depth, and height, respectively. A single sensitive region 151 was positioned so as to be embedded in the upper center of a virtual block whose width, depth, and height were 20 μm, 10 μm, and 10.25 μm, respectively. Then, the transistor layer 150 was set up by arranging these virtual blocks horizontally in a two-dimensional manner in the region spanning the substrate layer 110 and the insulating layer 120. The transistor layer 150 was positioned within the structural model 100 such that the boundary between the upper rectangular parallelepiped and the lower rectangular parallelepiped coincided with the boundary between the substrate layer 110 and the insulating layer 120.

[0055] To set the particle concentration in the encapsulant, a method was introduced to arrange the particles within the encapsulant layer 140 to satisfy a specified concentration. Figure 5 shows this method. In this method, a cube 141 containing four particles 142 with a diameter of 10 μm was used as the constituent unit of the encapsulant layer 140. The length of one side of the cube 141 could be changed between 20 μm and 100 μm. In state ST1 shown in Figure 5, the length of one side of the cube 141 is 20 μm, and in state ST2, the length is 100 μm. The larger the dimensions of the cube 141, the lower the particle concentration in the encapsulant.

[0056] The arrangement of the four particles 142 within the cube 141 is as follows: Two particles 142 were placed near the bottom surface of the cube 141 along one diagonal of a virtual cross-section obtained by virtually cutting the cube 141 horizontally. The remaining two particles 142 were placed near the top surface of the cube 141 along the other diagonal of the same virtual cross-section. The particles 142 were arranged within the cube 141 such that the distance between them increases as the dimensions of the cube 141 increase.

[0057] (Cosmic Ray Simulation) A cosmic ray simulation using structural model 100 was conducted under conditions simulating the International Space Station (ISS) located at an altitude of 400 km. PHITS (Particle and Heavy Ion Transport Code System) was used to perform this cosmic ray simulation. PHITS is a Monte Carlo calculation code that simulates various radiation behaviors in materials using nuclear reaction models, nuclear data, etc. In cosmic ray simulations, １ H, ４ He, ７ Li, ９ Be, １１ B, １２ C, １４ N, １６ O, ２０ Ne, ２４ Mg, ２８ Si, and ５６ Multiple cosmic rays, each originating from a different Fe nucleus, were virtually irradiated from above the structural model 100. The energy range of the cosmic rays was set to 10 －７ ~10 ３ The Mev value was set, and while randomly determining the energy value of cosmic rays within this energy range, the process of irradiating the structural model 100 with 10,000 cosmic rays was repeated 10,030 times.

[0058] (Generation of the first profile) The number of cosmic rays absorbed in one second by 3200 sensitive regions 151 (transistors) located in the central part of the transistor layer 150 was obtained as the number of events. The energy value of each cosmic ray absorbed by these 3200 sensitive regions 151 was also obtained. Using this obtained data, a first profile was generated showing the number of events for each of several energy values.

[0059] (Conversion to the second profile) The generated first profile was converted to a second profile that shows the number of events for each of the multiple charge quantities. For this conversion, a conversion constant obtained by the following formula was used: (Conversion constant) = (Elementary charge) / (Energy value required to generate an e-h pair) As mentioned above, silicon was assumed to be the material of the transistor, so the conversion constant is 1.6 × 10⁻⁶ －１９ [C] / 3.6 [eV] = 44.5 [fC / MeV]. Each energy value shown in the first profile was converted to an electric charge by multiplying it by its conversion constant, thereby converting the first profile to the second profile.

[0060] (Circuit Simulation) As a circuit simulation, the operation of a count-up circuit was simulated. For the circuit simulation, the specifications of the transistor were set as the design values ​​of the MOSFET: channel width, channel length, mobility, gate oxide capacitance, and threshold voltage. For the D-type flip-flop, the interval and frequency of the square wave, the frequency of the clock signal, and the delay time of the input signal and the clock signal were set. Furthermore, as semiconductor properties, the operating temperature, band gap, and doping level were set based on the Boltzmann distribution. In the circuit simulation, forward current was repeatedly output from the transistor at predetermined time intervals based on these settings.

[0061] The second profile obtained by converting the first profile shows the total for 3200 transistors. By dividing the number of events corresponding to each charge amount shown in the second profile by 3200, a second profile for one transistor was obtained as a unit second profile. Using this unit second profile, error currents were repeatedly generated in synchronization with the forward current. Each error current is a minute current obtained by dividing a randomly selected charge amount from the unit second profile by a predetermined minute time. Here, "predetermined" means that any minute time can be set.

[0062] (Identification of Soft Errors) In the circuit simulation, the forward current and error current generated simultaneously were compared, and when the error current was greater than or equal to the forward current, it was identified as a soft error. This comparison was repeated to obtain the number of soft errors that occurred per second. The number of soft errors that occurred per second was then calculated as 10 ９ The number of errors per unit of time was converted to this value, and this converted value was obtained as the soft error rate.

[0063] Gd ２ O ３ Sm ２ O ３ For each of the above procedures, and BN, by repeatedly performing the above series of steps while changing the particle concentration (mass % or volume %), we were able to verify the range of concentrations that control or reduce the effects of cosmic rays on electronic components. The results obtained are shown in Table 1. Here, "relative error rate" is the ratio of these particles (Gd ２ O ３ Sm ２ O ３ The rate of soft errors C when not including , and BN) ０ The ratio of the soft error rate C when particles are present to (C / C) ０ ) refers to.

[0064]

[0065] 10...Simulation system, 11...Model generation unit, 12...Cosmic ray simulation unit, 13...Conversion unit, 14...Circuit simulation unit, 100...Structural model, 110...Substrate layer, 120...Insulating layer, 130...Metal layer, 140...Sealing material layer, 141...Cube, 142...Particle, 150...Transistor layer, 151...Sensitive region.

Claims

1. A simulation system comprising at least one processor, wherein the at least one processor obtains a first profile indicating the number of events per unit time in which a cosmic ray having a certain energy value is absorbed by a transistor for each of a plurality of energy values ​​constituting the energy distribution of cosmic rays, converts each of the plurality of energy values ​​into a charge quantity, converts the first profile into a second profile indicating the number of events for each of the plurality of charge quantities, performs a circuit simulation which involves repeatedly comparing the forward current from the transistor based on the specifications of the transistor with the error current determined based on the second profile, and identifies a soft error in the circuit simulation when the error current is greater than or equal to the forward current.

2. The simulation system according to claim 1, wherein at least one processor generates a structural model representing an electronic component comprising a encapsulant containing a plurality of particles having the ability to absorb cosmic rays and a transistor protected by the encapsulant, wherein a encapsulant layer representing the encapsulant is stacked on top of a transistor layer representing the transistor; performs a cosmic ray simulation in which the plurality of cosmic rays are virtually irradiated onto the structural model with the plurality of cosmic rays passing through the encapsulant layer and the transistor layer in that order; generates the first profile based on data showing the cosmic rays absorbed by the transistor in the structural model in the cosmic ray simulation; and obtains the generated first profile.

3. The simulation system according to claim 2, wherein at least one processor receives user input specifying the concentration of the plurality of particles in the encapsulant, and generates the encapsulant layer of the structural model such that the arrangement of the plurality of particles in the encapsulant satisfies the concentration.

4. The simulation system according to claim 2 or 3, wherein at least one processor generates the structural model to represent the plurality of particles composed of one of gadolinium oxide, samarium oxide, and boron nitride.

5. The at least one processor, １ H, ４ He, ７ Li, ９ Be, １１ B, １２ C, １４ N, １６ O, ２０ Ne, ２４ Mg, ２８ Si, and ５６ Fe, each using respective atomic nuclei thereof as radiation sources, virtually irradiates the structural model with the plurality of cosmic rays, the simulation system according to claim 2 or 3.

6. A simulation method performed by a simulation system comprising at least one processor, comprising: obtaining a first profile indicating the number of events per unit time in which a cosmic ray having a certain energy value is absorbed by a transistor for each of a plurality of energy values ​​constituting the energy distribution of a cosmic ray; converting each of the plurality of energy values ​​into a charge quantity and converting the first profile into a second profile indicating the number of events for each of the plurality of charge quantities; performing a circuit simulation in which a comparison is made multiple times between a forward current from the transistor based on the specifications of the transistor and an error current determined based on the second profile; and identifying a soft error in the circuit simulation when the error current is greater than or equal to the forward current.

7. A simulation program that causes a computer to perform the following steps:

7. Obtain a first profile showing the number of events per unit time, which is the number of times a cosmic ray having a given energy value is absorbed by a transistor, for each of a plurality of energy values ​​that constitute the energy distribution of cosmic rays; convert each of the plurality of energy values ​​into a charge quantity, and convert the first profile into a second profile showing the number of events for each of the plurality of charge quantities; perform a circuit simulation that repeatedly compares the forward current from the transistor based on the specifications of the transistor with the error current determined based on the second profile; and identify a soft error in the circuit simulation when the error current is greater than or equal to the forward current.

Images