JP2010062255A - Method of analyzing semiconductor device and semiconductor analyzer - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To efficiently perform the optimization of process conditions such as a wiring process and a gate process in a short period of time by performing the dielectric constant evaluation of an interlayer film with high precision in the middle of a semiconductor manufacturing process. <P>SOLUTION: By applying a dielectric constant measuring method using electron energy loss spectroscopy to a semiconductor device in the middle of a process, the dielectric constant in a localized region is evaluated. Moreover, the intake angle and the sample film thickness at the time of EELS spectrum acquisition are measured, and the accurate intake angle and film thickness are used to perform the treatment of a spectrum and dielectric constant evaluation with high precision. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an analysis technique for a semiconductor device, and more particularly to a technique effective for evaluating a dielectric constant of an interlayer film or the like in a semiconductor device in the middle of a manufacturing process.

With the increase in speed and integration of semiconductor devices, for example, copper (Cu) wiring having a lower resistance than aluminum (Al) and SiOC interlayer film (Low-k film) having a lower dielectric constant than SiO ₂ have been used. Yes.

  However, the low-k film has been found to have a problem that the dielectric constant increases after the processing due to ashing damage during the processing, or the dielectric constant increases due to diffusion of Cu element into the low-k film.

  As a technique for measuring the dielectric constant, measurement using light such as Fourier transform infrared spectroscopy is known. However, since the probe size of light is on the order of micrometers, evaluation in a local region on the order of nanometers in a semiconductor device is not possible.

  An electron energy loss spectroscopy (EELS) using an electron microscope capable of reducing the probe size to the nanometer order is known as a technique for measuring the dielectric constant of a minute region.

  The spectrum of EELS is classified into two types: a low energy (up to several tens of eV) valence electron excitation spectrum (low loss spectrum) and a high energy (up to several keV) inner shell electron excitation spectrum (core loss spectrum).

Since the core loss spectrum by inner-shell electron excitation has a peak in energy intrinsic to the element, it is used for elemental analysis (see, for example, Patent Document 1). A low-loss spectrum is a loss spectrum of electron energy due to excitation of valence electrons in a solid, and it is known that a dielectric function is obtained by converting the low-loss spectrum to Kramers-Kronig (KK) (for example, non-loss spectrum). Patent Document 1).
JP 2004-265879 A Egerton (Author), Electron Energy Loss Spectroscopy in the Electron Microscope, 1986, pp. 197- 241-247, ISBN 0-306-42158-5

  However, the present inventors have found that the dielectric constant measurement technique as described above has the following problems.

  That is, as described above, when the dielectric constant is evaluated using light, the probe size of the light is on the order of micrometers, so that there is a problem that the local region on the order of nanometers cannot be evaluated.

  In the method of measuring the dielectric constant using electron energy loss spectroscopy (EELS) using an electron microscope capable of obtaining a nanometer-sized beam diameter, if there is physical information of the refractive index, the local region Although the dielectric constant could be measured, there is a problem that the dielectric constant of a material that is damaged and whose refractive index is unknown cannot be measured.

  Furthermore, in the method of measuring the dielectric constant of an unknown substance using EELS, in order to accurately process the spectrum, it is important to grasp the spectrum capture angle and the electron beam irradiation angle to the sample. As the lens conditions change, the spectrum capture angle and the electron beam irradiation angle change, which causes a problem that the measurement accuracy is lowered.

  In addition, in the technology for evaluating the dielectric constant using the spectral intensity of EELS, even if it is an interlayer film having the same dielectric constant, the film thickness cannot be accurately measured. Therefore, there is a problem that the dielectric constant cannot be accurately evaluated.

  An object of the present invention is to provide a technique for efficiently and quickly optimizing process conditions such as a wiring process and a gate process by accurately evaluating a dielectric constant of an interlayer film during a semiconductor manufacturing process. is there.

  The above and other objects and novel features of the present invention will be apparent from the description of this specification and the accompanying drawings.

  Of the inventions disclosed in the present application, the outline of typical ones will be briefly described as follows.

  The present invention includes a step of irradiating a specimen to be inspected with an electron beam, a step of detecting an electron beam transmitted through a portion irradiated with the electron beam for each energy, and a step of displaying the intensity of the detected electron beam for each energy. A step of monitoring the lens current when the electron beam is irradiated, a step of irradiating the standard sample with a known lattice spacing, a step of detecting a diffraction pattern formed by electrons transmitted through the standard sample, Using the diffraction pattern to calculate the capture angle, calibrating the lens current and capture angle, and correcting the capture angle using the calibration result from the lens current value monitored when the specimen was irradiated with the electron beam A step of correcting the intensity of the spectrum measured by irradiating the specimen to be inspected with an electron beam based on the calculated capture angle, and the intensity of the total energy of the spectrum with the capture angle corrected Measured by calculating the film thickness at the evaluation location by dividing the sum by the sum of the intensity of the zero-loss peak where no energy was lost and multiplying the logarithm of the result of division by a reference constant A step of correcting the spectrum by the film thickness, a step of evaluating the dielectric constant of the sample to be inspected using the spectrum of which the film thickness has been corrected, and moving the sample to irradiate an electron beam to multiple points of the sample or deflecting the electron beam The spectrum obtained while irradiating the sample is corrected with a series of capture angles and film thicknesses, and the dielectric constant is evaluated for each irradiation region.

  Further, the present invention includes a step of displaying the dielectric constant distribution of the sample to be inspected by converting the dielectric constant of the sample to be inspected into luminance and mapping it.

  Furthermore, the present invention provides an electron spectrometer that splits an electron beam for each energy, an electromagnetic lens that is provided downstream of the electron spectrometer and corrects aberrations of the electron spectrometer, and between the electron spectrometer and the electromagnetic lens. And a detector for detecting an electron intensity distribution.

  The present invention also includes a deflector provided between the detector and the electron spectrometer so that the detector can detect an electron beam, and deflects the electron beam, and the detector does not interact with the sample. Detects the electron intensity distribution on the focal plane of the electron spectrometer in addition to the optical path passing through the electron spectrometer and passing through the electromagnetic lens downstream of the electron spectrometer.

  Furthermore, the present invention displays the current value or the capture angle of the electromagnetic lens on a monitor, and displays a warning on the monitor when the current value or the capture angle of the electromagnetic lens is greater than or less than a desired value. Then, the control unit for controlling the analysis apparatus is provided with a control means for interrupting the analysis or setting and re-measuring the lens current value or the capture angle to be a desired value.

  Moreover, the outline | summary of the other invention of this application is shown briefly.

  In order to evaluate the dielectric constant of the local region, electron energy loss spectroscopy using an electron microscope capable of narrowing the beam diameter to a nanometer size is used.

  In addition, since the dielectric constant evaluation method by EELS is applied to a sample whose refractive index is unknown, the electron scattering formula of Formula 2 shown in Embodiment 1 is used without using a processing method using the refractive index in spectrum normalization processing. Is used to perform spectrum normalization processing.

  Further, in order to accurately process the EELS spectrum, the spectrum capture angle is measured as follows.

  Using a standard sample with a known lattice spacing, record the lens current value and diffraction pattern when the lens current of the electron microscope device is continuously changed, calculate the capture angle from the diffraction pattern, and capture for each lens current value Stores the angle, then stores the lens current value when the spectrum is measured by irradiating the part of the interlayer film analyzed by the semiconductor device to be inspected, and the lens current value calculated using the standard sample The capture angle at the time of spectrum measurement of the inspected part is calculated from the measurement data of the spectrum capture angle for each and the current value evaluated for the interlayer film of the semiconductor device to be inspected.

  In addition, the irradiation angle of the electron beam to the sample is set to about 0.5 mrad or less by using a condenser lens and a condenser aperture so that the angle at which the electron beam is irradiated on the sample does not affect the spectrum capturing angle. .

  Furthermore, regarding the measurement of the sample film thickness, the film thickness is measured for each evaluation location using the values obtained by measuring the overall intensity of the low loss spectrum of EELS and the intensity of the zero loss spectrum and Formula 5 shown in Embodiment 1 described later. It is to be measured.

  Among the inventions disclosed in the present application, effects obtained by typical ones will be briefly described as follows.

  (1) It becomes possible to correct the variation in the spectral intensity of EELS accompanying the apparatus fluctuation of the semiconductor analyzer, and to evaluate the dielectric constant of a substance whose refractive index is unknown with high resolution and high accuracy.

  (2) In addition, by applying the dielectric constant evaluation technology for interlayer film of semiconductor devices to semiconductor process development, the dielectric constant characteristics of semiconductor devices can be evaluated with high accuracy during the manufacturing process, improving the development efficiency of semiconductor products And manufacturing cost can be reduced.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. Note that components having the same function are denoted by the same reference symbols throughout the drawings for describing the embodiment, and the repetitive description thereof will be omitted.

(Embodiment 1)
1 is an explanatory diagram of a dielectric constant measurement technique according to Embodiment 1 of the present invention, FIG. 2 is a flowchart showing an example of a general dielectric constant evaluation method, and FIG. 3 is according to Embodiment 1 of the present invention. FIG. 4 is a block diagram showing a configuration example of a semiconductor analyzer for evaluating the dielectric constant of the semiconductor device according to the first embodiment of the present invention, and FIG. 5 is a low-loss spectrum scattering. FIG. 6 is an explanatory diagram showing an example of operation of the semiconductor analysis apparatus of FIG. 4.

  A technique for analyzing the dielectric constant of an interlayer film in a semiconductor device in which an interlayer film is formed during the semiconductor manufacturing process in the first embodiment will be described.

  First, the cause of the decrease in measurement reproducibility when obtaining a dielectric function using electron energy loss spectroscopy (EELS) will be described.

  A flow until obtaining the dielectric function shown in Non-Patent Document 1 will be described using EELS in FIGS. 1 and 2.

  First, the background of the zero-loss peak where no energy is lost is removed from the measured low-loss spectrum shown in FIG. 1A (step S001). Next, since the spectrum obtained in the process of step S001 contains multiple scattering components, the multiple scattering components are removed, and the single scattering spectrum S (ω) shown in FIG. Derived (step S002).

  A Fourier log method is used to remove multiple scattering components in the low-loss spectrum. Then, Im [−1 / ε (ω)] called a loss function shown in FIG. 1D is derived from the single scattering spectrum (step S003).

  Here, ε (ω) represents the dielectric function of Equation 1, and Im [−1 / ε (ω)] represents the imaginary part of −1 / ε (ω). Finally, using the loss function (Im [−1 / ε (ω)]) K. By performing the conversion, the dielectric function shown in FIGS. 1E and 1F is obtained (step S004).

  Next, a method for deriving the loss function from the single scattering spectrum in step S003, which is a factor that hinders measurement reproducibility, will be described. Each processing method of steps S001, S002, and S004 has been established and is omitted here. Please refer to Non-Patent Document 1 for the detailed processing method.

  The loss function (Im [−1 / ε (ω)]) (FIG. 1 (d)) is a loss function in the intensity of the single scattering spectrum S (ω) (FIG. 1 (c)) as shown in Equation 2. Are normalized by Ap (ω) called an aperture function. The absolute value k of the loss function and the aperture function Ap (ω) are expressed as in Expression 3 and Expression 4, respectively.

Where β is the spectrum capture angle, θ _E is the characteristic scattering angle, a ₀ is the Bohr radius, m ₀ is the electron's stationary mass, v is the velocity of the incident electron, I ₀ is the intensity of the zero-loss beam, and t is the sample. Represents the film thickness.

  The aperture function Ap (ω) and the absolute value k of the loss function are obtained from the experimental conditions and Equations 3 and 4, and then the loss is obtained from the single scattering spectrum (FIG. 1 (c)) by using Equation 2. A function (FIG. 1 (d)) is obtained.

  The absolute value k of the above-described loss function includes an equation using a refractive index in addition to Equation 3, but the description is omitted here because it cannot be applied to a substance whose refractive index is unknown.

  Next, the cause of reducing the measurement reproducibility will be described.

  In the derivation of the loss function, Equations 2 to 4 are used. From the parameters included in Equations 2 to 4, the factors causing the decrease in measurement reproducibility are the spectrum capture angle (β) and the film thickness (t) of the sample.

Another factor is the variation in the electron velocity incident on the sample. The variation in the incident electron velocity is caused by the variation (ΔV ₀ ) in the acceleration voltage (V ₀ ), and the variation in the acceleration voltage (ΔV ₀ / V ₀ ) is as small as 10 ⁻⁶ , so the variation in incident electron velocity can be ignored.

  That is, in order to improve the measurement reproducibility, it is important to accurately grasp the spectrum capture angle (β) and the film thickness (t) of the sample at the time of spectrum measurement. FIG. 3 shows the calculation result of the amount of change in dielectric constant when the spectrum capture angle (β) and the film thickness (t) of the sample are changed.

  The calculation results in FIG. 3 are obtained by using the same single scattering spectrum and deriving a plurality of loss functions by independently changing the parameters of the capture angle (β) and the film thickness (t) of the sample. The loss function K. It is obtained by conversion processing.

  From the result of dependence on the measurement angle (β) of the variation in dielectric constant measurement (FIG. 3A), it was predicted that when the standard 4 mrad was changed by ± 1 mrad, the fluctuation was about 5 to 10%.

  From the result of the dependence of the variation in dielectric constant measurement on the film thickness (t) (Fig. 3 (b)), it is expected that the dielectric constant will vary by 5% when there is a 5% error in the film thickness measurement of the sample. It was. For this reason, in order to set the dielectric constant measurement variation by EELS to 5% or less, it is necessary to set the spectrum capture angle (β) to 1 mrad or less and the measurement error of the sample film thickness to 5% or less.

  Next, FIG. 4 shows the configuration of the semiconductor analysis apparatus 1 that performs analysis of the semiconductor device according to the present embodiment.

  The electron beam 2 emitted from the electron gun 1 a passes through the sample 5 held on the condenser lens 3, the condenser diaphragm 4, and the sample stage 6, and forms an image at a position upstream of the intermediate lens 9 by the objective lens 7.

  Although the objective lens 7 is simply illustrated as a single lens, the objective lens 7 is actually composed of an upper magnetic pole lens and a lower magnetic pole lens, and the objective diaphragm 8 located downstream of the sample 5 and the sample 5 is the objective lens 7. The upper magnetic pole lens is installed between the lower magnetic pole lens and the lower magnetic pole lens.

  The objective aperture 8 is installed at a position where the rear focal plane of the objective lens 7 is formed. Using the current condition of the condenser lens 3 and the condenser diaphragm 4 so that the electron beam spread (beam diameter) of the electron beam 2 irradiated onto the sample 5 is nanometer size and the irradiation angle of the electron beam 2 is 0.5 mrad or less. Yes.

  FIG. 5 shows the angular distribution of low-loss electrons when the irradiation angle of the electron beam is changed.

  When the irradiation angle of the electron beam is 0 mrad, the spread of the low-loss electrons is such that the scattering angle of the low-loss electrons is about 0.1 mrad with respect to the traveling direction of the incident electrons, and the incident angle of the electron beam is several mrad. The scattering angle of the low-loss spectrum is also several mrad.

  The reason why the irradiation angle of the electron beam is set to 0.5 mrad or less is to eliminate the change in the intensity of the spectrum due to the irradiation angle because the capture angle of the spectrum is several mrad. An image formed by the objective lens 7 of the sample is magnified by the intermediate lens 9 and the projection lens 10 and detected by the detector 11.

  The electron beam 2 transmitted through the sample 5 by moving the detector 11 passes through the objective lens 7, the intermediate lens 9, and the projection lens 10, and is dispersed for each energy by the electron spectrometer 13 and detected by the EELS detector 15. Is done.

  An electromagnetic lens 12 for correcting the aberration of the electron spectrometer 13 is upstream of the electron spectrometer 13, a detector 34 for detecting the electron intensity distribution of the focal plane is downstream of the electron spectrometer 13, and electron spectroscopy. An electromagnetic lens 14 is provided for focusing and projecting the focus of the device 13 on the EELS detector 15.

  The detector 34 is configured to be installed on the optical axis or other than the optical axis. When the detector 34 is installed on the axis, the detector 34 has a detector driving mechanism 36 for removing from the optical axis at the time of spectrum measurement. The detector 34 has a configuration in which the deflector 35 is installed so that the focal plane enters the detector 34 when the detector 34 is installed other than on the optical axis.

  The electromagnetic lens 14 is simply a single lens, but has a four-stage lens configuration for focus adjustment and enlargement / reduction projection. This lens is adjusted by the operation unit 25 through the control means 16 described below.

  The control means 16 includes an electron gun control unit 17, a condenser lens control unit 18, a stage control unit 19, an objective lens control unit 20, an intermediate lens control unit 21, a projection lens control unit 22, an electron spectrometer aberration correction lens control unit 23, And an electron spectrometer focus adjustment lens control unit 24.

  The operation unit 25 includes an image display unit 26, a spectrum display unit 27, a lens current monitor unit 28, a storage unit 29 that stores lens current, images, and analysis data, an arithmetic processing unit 30, and an operation screen 31 shown in FIG. Is done.

  Next, the EELS spectrum capture angle will be described.

  The spectral capture angle 37, which is a solid angle seen from the EELS detector 15 of electrons emitted from the sample 5, is an objective diaphragm installed at the focal position of the objective lens 7 when the image plane of the sample is projected onto the detector 11. 8 is limited by the hole diameter of the objective aperture 8.

  Therefore, the spectrum capture angle is determined by detecting a diffraction image projected on the detector 11 using a sample having a known lattice interval, or a focal plane downstream of the electron spectrometer when an image plane is formed on the detector 11. It is possible to measure by using the detector 34 that performs the measurement or by using the EELS detector 15 after being reduced by the electromagnetic lens 14.

  Usually, when calculating the capture angle using the diffraction image detected by the detector 11, the diffraction image is recorded with a film or the like, then opened to the atmosphere, and a plurality of processes such as development are required. It takes time to read.

  On the other hand, when the detector 34 or the EELS detector 15 is used, the detected signal is displayed on the operation display unit 26, and the arithmetic processing unit 30 performs the arithmetic processing of Expressions 2, 3, and 4. It is possible to omit the steps of opening the vacuum system to the atmosphere and developing the film, and to measure the capture angle and the spectrum with high accuracy in a short time.

  When the lens condition of the semiconductor analyzer 1 changes, the distance between the diffraction points of the diffraction image at the position of the objective aperture 8 changes, but the hole diameter of the objective aperture 8 does not change physically, so the actual capture angle Will change.

  As a result, there is a problem in that the measurement accuracy decreases because the spectrum processing shown in Expression 2 and Expression 4 is not performed using an accurate spectrum capturing angle. In order to solve this problem, in the present invention, before and after acquiring a EELS spectrum by irradiating a semiconductor device with an electron beam, a spectrum capture angle is accurately measured using a diffraction pattern of a sample having a known lattice spacing.

  In addition, a spectrum capture angle for each lens current is obtained in advance from a diffraction pattern obtained by changing the lens current, stored in the storage unit 29, and the lens current when the spectrum is obtained by irradiating the electron beam to the semiconductor device is the lens current. The capturing angle at the time of spectrum measurement can be accurately measured from the relationship between the lens current value when the spectrum is acquired by monitoring by the monitor unit 28, the lens current value stored in advance and the capturing angle.

  Next, measurement of the film thickness (t) will be described.

FIG. 7 shows the entire low-loss spectrum 32 detected by the EELS detector 15, and the hatched portion is the zero-loss spectrum 33 where no energy is lost. By calculating the intensity I _total of the entire low-loss spectrum 32 and the intensity I ₀ of the zero-loss spectrum 33 and using Equation 5, the film thickness at the evaluation location can be measured.

  Λ in Equation 5 represents the mean free process, and in this embodiment, the value obtained from the reference sample was used. Usually, the specimen to be inspected for transmission electron microscope observation uses a thinned piece so as to have a uniform film thickness, but it is difficult to make the specimen to be examined containing a plurality of materials uniformly thin. Therefore, the variation in film thickness may be several tens of percent or more.

  In order to reduce the measurement variation due to the variation in film thickness to several percent or less, in the present invention, the arithmetic processing unit 30 uses the zero loss intensity of the low loss spectrum and the total intensity result of the low loss spectrum that are actually measured for each evaluation point. By performing arithmetic processing, the film thickness is accurately measured for each evaluation location.

  As a result, according to the first embodiment, it is possible to accurately measure the capture angle and film thickness at the time of spectrum measurement, and use the accurate capture angle and film thickness for each spectrum to be measured. By performing the processing, it is possible to evaluate the dielectric constant with high measurement accuracy.

(Embodiment 2)
FIG. 8 is a flowchart showing a processing example of the dielectric constant evaluation technique according to the second embodiment of the present invention, FIG. 9 is an explanatory diagram of a method for measuring the capture angle, and FIG. 10 is an explanation showing the relationship between the lens current and the capture angle FIG. 11 is an explanatory diagram of a dielectric constant evaluation technique according to the second embodiment of the present invention, FIG. 12 is an explanatory diagram of the dielectric constant evaluation technique according to the second embodiment of the present invention, and FIG. 13 is an implementation of the present invention. It is explanatory drawing which shows the result at the time of applying to the semiconductor device by Embodiment 2.

  In the second embodiment, the semiconductor analysis apparatus 1 (FIG. 1) of the first embodiment is used to accurately measure the spectrum capture angle and the sample film thickness, perform the spectrum processing, and are included in the semiconductor device. A technique for evaluating the dielectric constant of the interlayer film will be described.

  The inspection method will be described with reference to the flowchart of FIG.

  First, a calibration sample with a known lattice spacing is carried into the semiconductor analyzer 1 (step S101). Subsequently, adjustment of the electron optical system and EELS spectrum measurement conditions are set, and a spectrum capture angle is measured using a diffraction pattern (step S102).

  Here, an example of measuring the capture angle will be described.

  For example, when a Si crystal sample is used as the calibration sample for the capture angle, the Si diffraction pattern 38 shown in FIG. 9 is detected by the detector 11 and displayed on the operation screen 31 (FIG. 6). The circular contour portion 39 shown in FIG. 9 is a shadow of the hole portion of the objective aperture 8, electrons outside the circular contour portion 39 are removed, and electrons passing inside the contour portion 39 are removed from the EELS detector. 15 is detected.

  The detected electron capture angle can be measured by the following procedure.

First, an electron beam when the distance R ₁ 42 between the 0th-order diffraction point 40 and the (111) diffraction point 41 in the diffraction pattern 38, the inter-surface distance d ₁₁₁ = 3.136 mm of Si (111), and the acceleration voltage is 300 kV. The camera length L is measured using Equation 6 from the wavelength λ = 1.97 pm.

Then, from the camera length L obtained earlier, the distance R ₀ 43 from the 0th-order diffraction point 40 in the diffraction pattern 38 to the contour portion 39 of the objective aperture, and the electron beam wavelength λ = 1.97 pm, d ₀ is used. Is calculated.

Then, θ is obtained by using Equation 7 from d ₀ and the electron beam wavelength λ = 1.97 pm. This θ is the spectrum capture angle (β) limited by the objective aperture 8. In the apparatus of the present invention, based on the conditions input on the input screen 45 provided on the operation screen 31 shown in FIG. The capture angle (β) is output and displayed on the top.

That is, in the input screen 45 provided on the operation screen 31, the input field 46 of the acceleration voltage (or electron beam wavelength), the input field 47 of the diffraction point distance R ₁ 42 corresponding to the known surface interval, and the diffraction point distance are supported. Simply enter each value in the input field 48 for the lattice spacing d and the input field 49 for the distance R ₀ 43 from the 0th-order diffraction center to the contour portion 39 corresponding to the shadow of the hole portion of the objective aperture 8. The capture angle (β) is displayed on the screen 31.

For this reason, the user of the electron microscope does not have to perform a plurality of calculations, which is efficient and easy to use. The diffraction point distance R ₀ 43 described above is configured such that the diffraction processing unit 30 calculates and outputs the diffraction point interval distance by designating the position of each diffraction point on the diffraction pattern 38. The distance between the diffraction spots may be an average of the distances of two or more diffraction spots on the straight line.

  Next, the capture angle is measured using a diffraction image obtained by changing the lens conditions, and the capture angle for each lens current is stored in the storage unit 29 (step S103). The capture angle is measured by the same method as that in step S102.

  Then, based on the relationship between the capture angle and the current value stored in the storage unit 29 shown in FIG. 10, an output display 50 for the lens current value and an output display 51 for the capture angle are provided on the operation screen 31 of the operation unit. Display each one.

  The sample for calibration of the spectrum capture angle used in the process of step S103 may be a Si substrate included in the semiconductor device to be inspected similar to the process of step S102, or a sample that is constantly held on the sample stage 6.

  Next, the calibration sample is carried out from the inspection apparatus, the semiconductor device to be inspected is carried into the semiconductor analysis apparatus 1, and the above-described electron optical adjustment is performed again (step S104). When the Si substrate included in the sample to be inspected is used, or when the taking-in angle is measured using a sample substrate that has been previously installed, the process of step S104 can be omitted.

  Subsequently, the electron beam 2 is irradiated onto the semiconductor sample 5, the electrons emitted from the sample are dispersed by the electron spectrometer 13, and the low loss spectrum is detected by the EELS detector 15. The lens current at the time of acquiring the low-loss spectrum is monitored by the lens current monitor unit 28 and stored in the storage unit 29. The irradiated part is also stored in the storage unit 29 at the same time.

  The portion (Xn, Yn) 53 where the sample 5 to be inspected is irradiated with the electron beam 2 is scanned by using the control means 16 of the deflector 54 for deflecting the electron beam shown in FIGS. Use and memorize.

  The series of spectrum acquisition, lens current monitoring, and lens current storage are performed while deflecting the electron beam or moving the sample stage 6 (step S105). If the lens current value deviates from the desired condition range during the spectrum measurement, a warning 52 is displayed on the operation screen 31 to stop the spectrum measurement, set the optical condition to the desired lens current condition, and Perform re-measurement.

  Next, a spectrum capture angle at the time of evaluation is calculated from the lens current value monitored for each evaluation location, the processing in step S102, and the spectrum calibration data obtained in the processing in step S103 (step S106). The arithmetic processing unit 30 performs arithmetic processing using the low-loss spectrum data obtained for each time and Equation 5, and calculates the film thickness for each evaluation location (step S107).

  Further, a specific example of the processing in steps S106 and S107 will be described with reference to FIG.

  A plurality of spectrum measurement results 55 in FIG. 11 indicate the spectrum for each portion irradiated with the electron beam obtained in the processes of steps SS102 to S104 and the lens current value monitored by the lens current monitor unit 28. Is remembered.

The spectrum measurement result 55-n represents a spectrum and a lens current value obtained by irradiating the electron beam 2 with the location (Xn, Yn) 53 of the sample 5 to be inspected. The capture angle (β _n ) when the spectrum 55-n is measured is the relationship between the lens current value (I _n ) at the time of spectrum measurement and the spectrum capture angle obtained by the processing in step S103 and the lens current value (FIG. 10).

  As described in the first embodiment, the film thickness measurement in the process of step S107 calculates the intensity of the zero-loss spectrum and the low-loss spectrum intensity of the measured spectrum 55-n. It is calculated | required by carrying out 5 numerical formula processing.

  By performing the processing of steps S106 and S107 for each measured spectrum, it is possible to obtain the capture angle and film thickness value 56 for each measured spectrum.

  Next, the processing unit 30 removes the zero-loss spectrum from the low-loss spectrum (Step S108), and derives a single scattering spectrum (Step S109). In the derivation of the single scattering spectrum, the arithmetic processing unit 30 performs arithmetic processing using the Fourier log method.

  Thereafter, the single scattering spectrum 57 for each evaluation point obtained by the processing of steps S106 to S109 shown in FIG. 12 is expressed by using the capturing angle (β) and the film thickness (t) at the time of spectrum acquisition, using Expressions 2 and 3 , And Expression 4 are performed by the arithmetic processing unit 30.

  Thereby, the loss function Im [−1 / ε (ω)] with high measurement accuracy can be obtained for each evaluation point (step S110). Then, the arithmetic processing unit 30 analyzes the obtained loss function Im [−1 / ε (ω)] for each evaluation point, and obtains a real part 58 of the dielectric function for each evaluation point and an imaginary number of the dielectric function. Part 59 is derived.

Here, the processing used for the calculation processing of analysis includes K.I. K. (Kramers Kronig) analysis process was used. The value (ε ₁ (0)) 60-n of the real part of the dielectric function when the energy of the real part 58-n of the dielectric function at the evaluation place (Xn, Yn) is 0 is stored for each evaluation place, and brightness conversion is performed. Are displayed (step S111).

For example, in the region of the low-k material portion 62 formed between the Si substrate 61 and the SiO ₂ 63 of the sample to be inspected shown in FIG. As shown in FIG. 13B, a dielectric constant distribution 64 in the low-k material region can be obtained.

FIG. 13C shows the luminance display 65 of the dielectric constant 66. With reference to FIG. 13C, the dielectric constant is 4. in the vicinity of SiO ₂ in the Low-k material of FIG. 13B. You can see at a glance that it is close to zero.

  Here, the luminance display of the dielectric constant distribution in FIG. 13 shows an example in which the dielectric constant value is displayed as luminance. However, the change in the dielectric constant from the reference dielectric constant value is converted into luminance and displayed. Also good. The magnitude and direction of the dielectric constant may be displayed using the length (thickness) and direction of the arrow.

(Embodiment 3)
FIG. 14 is an explanatory diagram showing a result obtained by applying to a semiconductor device having different process conditions according to the third embodiment of the present invention.

  In the third embodiment, an example will be described in which the first and second embodiments are applied during the manufacturing process of the semiconductor device and fed back to the process of the semiconductor manufacturing development stage.

  Here, as described below, an example in which the present invention is applied to a low-k material in an interlayer film forming process is shown. However, the present invention is not limited to an interlayer film using a low-k material, and the high- It is also applicable to k material and other materials.

  In order to compare the performance of plasma-treated semiconductor devices using three types of plasma ions (plasma ion species A, plasma ion species B, and plasma ion species C), a wafer was extracted from the process line and this method was used. Was used to evaluate the dielectric constant of three types of semiconductor products.

  The measurement results of three types of semiconductor products obtained by this method are shown in FIG.

  FIGS. 14A to 14C show the dielectric constant distribution images of the sample to be inspected processed under the conditions of plasma ion species A, plasma ion species B, and plasma ion species C, and the low-k material part, respectively. Yes.

  14A to 14C show Si substrates 61a, 61b, and 61c in the sample to be inspected, respectively, and above these Si substrates 61a, 61b, and 61c, Low in the sample to be inspected. The dielectric constant distributions 64a, 64b, and 64c of the -k material portion are shown.

Further, the dielectric constant distribution 64a, 64b, above the 64c shows SiO ₂ 63a of the test sample, 63b, a 63c respectively. FIG. 14D shows a dielectric display 66-1 with a dielectric constant 66-1.

  From the measurement result of FIG. 14, the result that the dielectric constant increased in the order of the plasma ion species A, B, and C conditions was obtained. From the measurement result of the dielectric constant distribution under the plasma ion species A, B, and C conditions, the plasma treatment process using the plasma ion species A was determined to be optimal.

  Accordingly, in the third embodiment, by applying this method in the middle of the process of manufacturing the semiconductor device, it becomes possible to feed back to the process of developing the semiconductor device in a short period of time, thereby shortening the development period of the semiconductor device. be able to.

  As mentioned above, the invention made by the present inventor has been specifically described based on the embodiment. However, the present invention is not limited to the embodiment, and various modifications can be made without departing from the scope of the invention. Needless to say.

  The present invention is suitable for a technique for evaluating a dielectric constant of an interlayer film or the like in a semiconductor device in the middle of a manufacturing process.

It is explanatory drawing of the dielectric constant measurement technique by Embodiment 1 of this invention. It is a flowchart which shows an example of a dielectric constant evaluation method. It is explanatory drawing of the variation factor of the dielectric constant measurement by Embodiment 1 of this invention. It is a block diagram which shows the structural example of the semiconductor analyzer which performs the dielectric constant evaluation of the semiconductor device by Embodiment 1 of this invention. It is explanatory drawing of the scattering angle distribution of a low-loss spectrum. FIG. 5 is an explanatory diagram illustrating an operation example of the semiconductor analysis apparatus of FIG. 4. It is explanatory drawing of the film thickness measurement technique by Embodiment 2 of this invention. It is a flowchart which shows the process example of the dielectric constant evaluation technique by Embodiment 2 of this invention. It is explanatory drawing of the measuring method of a taking-in angle. It is explanatory drawing showing the relationship between a lens electric current and a taking-in angle. It is explanatory drawing of the dielectric constant evaluation technique by Embodiment 2 of this invention. It is explanatory drawing of the dielectric constant evaluation technique by Embodiment 2 of this invention. It is explanatory drawing which shows the result at the time of applying to the semiconductor device by Embodiment 2 of this invention. It is explanatory drawing which shows the result obtained by applying to the semiconductor device from which the process conditions differ by Embodiment 3 of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Semiconductor analyzer 1a Electron gun 2 Electron beam 3 Condenser lens 4 Condenser diaphragm 5 Inspected semiconductor sample 6 Sample stage 7 Objective lens 8 Objective diaphragm 9 Intermediate lens 10 Projection lens 11 Detector 12 Electromagnetic lens 13 Electron spectrometer 14 Electromagnetic lens 15 EELS detector 16 control means 17 electron gun controller 18 condenser lens controller 19 stage controller 20 objective lens controller 21 intermediate lens controller 22 projection lens controller 23 electron spectrometer aberration correction lens controller 24 electron spectrometer focus adjustment Lens control unit 25 Operation unit 26 Image display unit 27 Spectrum display unit 28 Lens current monitor unit 29 Storage unit 30 Calculation processing unit 31 Operation screen 32 Low loss spectrum 33 Zero loss spectrum 34 Detector 35 Deflector 36 Detector drive mechanism 37 Spectrum Capture angle 38 Diffraction pattern 9 contour portion 40 0th breakpoint 41 (111) diffraction spot 42 distance R ₁
43 Distance R ₀
45 Input screen 46 Input field 47 Input field 48 Input field 49 Input field 50 Output display 51 Output display 52 Warning display 53 Irradiation location 54 Deflector 55 Spectrum measurement result 55-n Spectrum measurement result 56 Spectrum measurement result 57 Single scattering spectrum 57 -N single scattering spectrum 58 real part 58-n real part 59 imaginary part 59-n imaginary part 60 real part value 60-n real part value 61 Si substrate 61a Si substrate 61b Si substrate 61c Si substrate 62 Low-k Material part 62a Low-k material part 62b Low-k material part 62c Low-k material part 63 SiO ₂
63a SiO ₂
63b SiO ₂
63c SiO ₂
64 Dielectric constant distribution 64a Dielectric constant distribution 64b Dielectric constant distribution 64c Dielectric constant distribution 65 Luminance display 65-1 Luminance display 66 Dielectric constant 66-1 Dielectric constant

Claims (5)

Irradiating the specimen to be inspected with an electron beam;
A step of detecting, for each energy, an electron beam transmitted through the portion irradiated with the electron beam;
Displaying the detected intensity of the electron beam for each energy;
Monitoring the lens current when irradiated with an electron beam;
Irradiating a standard sample with a known lattice spacing with an electron beam;
Detecting a diffraction pattern formed by electrons transmitted through the standard sample;
Calculating the capture angle using the diffraction pattern and calibrating the lens current and capture angle;
Correcting the capture angle using the calibration result from the lens current value monitored when the sample to be inspected is irradiated with an electron beam;
Correcting the intensity of the spectrum measured by irradiating the sample to be inspected with an electron beam based on the calculated capture angle;
Divide the sum of all the energy intensities of the spectrum corrected for the capture angle by the sum of the zero-loss peak intensities without energy loss, and multiply the logarithm of the result of division by a reference constant. A step of calculating the film thickness at the evaluation location;
Correcting the spectrum with the measured film thickness;
Evaluating the dielectric constant of the sample to be inspected using the spectrum with the corrected film thickness;
Move the sample and irradiate the sample with multiple electron beams or deflect the electron beam to irradiate the sample. And a step of evaluating the semiconductor device. In the analysis method of the semiconductor device according to claim 1,
A method for analyzing a semiconductor device, comprising: a step of converting a dielectric constant of the sample to be inspected into luminance and mapping it, and displaying a dielectric constant distribution of the sample to be inspected. An electron spectrometer that splits an electron beam for each energy;
An electromagnetic lens provided downstream of the electron spectrometer and correcting aberrations of the electron spectrometer;
A semiconductor analyzer comprising: a detector provided between the electron spectrometer and the electromagnetic lens and detecting an electron intensity distribution. The semiconductor analysis apparatus according to claim 3,
Provided between the detector and the electron spectrometer so that the detector can detect an electron beam, and comprising a deflector for deflecting the electron beam,
The detector is
A semiconductor analysis characterized in that an electron intensity distribution on the focal plane of the electron spectrometer is detected in addition to an optical path in which electrons that do not interact with a sample pass through the electron spectrometer and pass through the electromagnetic lens downstream of the electron spectrometer. apparatus. In the semiconductor analysis device according to claim 3 or 4,
The current value or the capture angle of the electromagnetic lens is displayed on the monitor, and if the current value or the capture angle of the electromagnetic lens is greater than or less than the desired value, a warning is displayed on the monitor to control the analysis device. A semiconductor analysis apparatus comprising: a control means for interrupting the analysis by the control unit or setting the re-measurement so as to obtain a desired lens current value or capture angle.

Images