JP7313499B2 - Scanning electron microscope and secondary electron detection method for scanning electron microscope - Google Patents

Description

The present invention relates to a scanning electron microscope and a secondary electron detection method for a scanning electron microscope that can prevent image SNR from deteriorating in a low-vacuum state used for antistatic and enable image acquisition with a higher SNR.

2. Description of the Related Art Electron microscopes are widely used at semiconductor device manufacturing sites to observe and measure semiconductor devices and the like with nanometer-order resolution. At a semiconductor device manufacturing site, a sample having a metal pattern provided on the surface of a silicon substrate, quartz constituting a photomask, or the like is mainly observed. Since these patterns are usually provided on an insulator, they are easily charged when irradiated with an electron beam, making correct observation and measurement difficult.

An electron beam is a stream of electrons, negatively charged particles, which scatter when they reach the sample surface, generating secondary electrons. The amount of secondary electrons generated from one primary electron varies depending on the energy of the sample and the primary electron beam, and varies within the range of 0 to 10.

Therefore, when the amount of primary electrons used for irradiation does not balance with the amount of secondary electrons generated, the sample surface is charged either positively or negatively. Since electrification occurs on the surface of the insulator, it remains in that state for a considerable period of time.

When charged, a large potential gradient is generated around it, which changes the trajectory and landing energy of the primary electron beam, resulting in a distorted image.

Since electrification also affects the trajectory of the generated secondary electrons, in the worst case, the image may become completely white or completely black, and nothing can be seen.

Conventionally, in order to observe patterns on insulators, various antistatic techniques have been used, such as the technique of measuring the potential of the observation point and applying the potential to electrodes placed in the vicinity of the observation point to keep the potential of the object to be measured constant, or the technique of scanning at high speed to prevent electrons from accumulating.

Among them, antistatic technology using positive and negative ion gas generated by interaction with an electron beam in a low-vacuum state has been found to be superior to other methods for observing partially insulated patterns such as photomasks and isolated island patterns.

Interaction with the electron beam (collision between the electron beam and the gas) is necessary to generate the ion gas, which reduces the amount of the primary electron beam and increases the beam spot size, resulting in deterioration of the image SNR and resolution.

Furthermore, secondary electrons generated on the surface of the sample interact with the ion gas on the way to the electron detector, disappear, do not reach the electron detector, and cannot be detected.

Furthermore, the more the ionizing gas is introduced into the vacuum chamber, the more the antistatic effect is improved, but the SNR of the acquired image is degraded.

In addition, secondary electrons or backscattered electrons generated on the surface of the sample interact with the ion gas present in the vacuum chamber to change their trajectory, and are scattered in all directions, failing to reach the electron detector. This lowers the detection efficiency of the secondary electrons.

If the SNR is lowered, there is a problem that it takes longer to acquire an image with the required SNR or it becomes impossible to acquire the image.

Conventionally, it was thought that a discharge would occur when a bias voltage was applied to a sample in a low vacuum, and in the current system, an interlock was applied to prevent the bias from being applied in the low vacuum state.

In the present invention, the secondary electrons generated on the surface of a sample (specimen) in a low vacuum are applied with an acceleration voltage within a range that does not discharge, and are accelerated in the direction perpendicular to the surface of the sample.

In addition, a separately provided ion generator is used to generate a large amount of ions that cannot be generated in a conventional chamber vacuum and discharge them toward a sample, thereby improving the image SNR while maintaining a sufficient antistatic effect while maintaining a high degree of vacuum in the vacuum chamber.

Therefore, the present invention provides a scanning electron microscope provided with a secondary electron detector for detecting emitted secondary electrons while irradiating a sample with a narrowly narrowed electron beam, wherein a small aperture through which the electron beam passes but hardly passes between an objective lens for narrowing the electron beam and a sample movably held in a sample chamber; A voltage applying device is provided, and the sample voltage applying device applies to the aperture a voltage that accelerates secondary electrons emitted from the sample toward the aperture, deflects the secondary electrons emitted from the sample so that they pass through a small hole in the aperture, and detects the passing secondary electrons by a secondary electron detector, thereby increasing the secondary electron detection ratio.

At this time, the predetermined pressure for maintaining the pressure in the sample chamber is set to a pressure at which the finely focused electron beam collides with the gas in the sample chamber to ionize the gas, and a sufficient amount of ions is obtained to neutralize the negative charge generated on the surface of the sample by the irradiation of the electron beam.

Further, the predetermined pressure for maintaining the pressure of the sample chamber at a predetermined pressure is set to a pressure at which ions from an ion generator connected to the sample chamber are introduced to obtain a sufficient amount of ions to neutralize the negative charges generated by the irradiation of the electron beam on the surface of the sample.

Also, in order to control the introduction of ions from the ion generator to the measurement point, a pulsed control voltage is applied to adjust the average ion flow.

Also, the predetermined bias voltage is set to 400 V or less at which discharge does not occur.

The present invention applies an acceleration voltage within a range that does not discharge secondary electrons generated on the surface of a sample in a low vacuum, accelerates the secondary electrons in a direction perpendicular to the surface of the sample, and makes the secondary electrons, which have conventionally been scattered in all directions, travel within the detection range of a secondary electron detector and makes them detectable.

In addition, by using an ion generator, a large amount of ions that cannot be generated in a conventional chamber vacuum are generated and emitted toward a sample, and the image SNR can be improved while maintaining a sufficient antistatic effect while maintaining a high degree of vacuum in the vacuum chamber. By improving the SNR, it is possible to improve the throughput of measurement accuracy and inspection speed. Since the irradiation current value can be lowered to obtain an image with the same SNR, damage to the sample can be reduced. Since the secondary electrons can be prevented from colliding with the aperture, charging of the aperture can be prevented and the electron beam irradiation axis can be stabilized.

FIG. 1 shows a block diagram of one embodiment of the present invention. FIG. 1 shows the basic configuration of an electron microscope utilizing low vacuum according to the present invention. The entire apparatus is automated so that it operates according to commands from an input device such as a keyboard while viewing a display 24 of a PC (personal computer) 23 . Communication between devices is performed using TCPIP, USB, or other serial communication means. The measurement conditions, results, etc. are stored in a database (not shown), and the optimum measurement conditions can be instantly read out and set, and the measurement results can be displayed on the display 24 for viewing.

In FIG. 1, an electron gun 1 is for generating an electron beam, and consists of a Schottky emitter with a tip of a tungsten crystal filled with ZrO or the like, an electrode such as an extractor for applying a strong electric field to the emitter, or a suppressor electrode for pushing in unnecessary electrons. The vicinity of the electron gun 1 is kept at a degree of vacuum higher than 10 to the minus 7th power in order to stably operate the Schottky type electron emitter.

The electron gun 1 portion and other portions are separated by two differentially configured ion pumps, and a high degree of vacuum is maintained in the electron gun 1 portion.

A condenser lens 2 collects the electrons emitted from the electron gun 1 and adjusts the electron density to an appropriate level.

The blanking electrode 3 is an electrode that applies a high voltage to irradiate the sample 11 with the electron beam only when necessary.

The blanking aperture 4 is for blocking the electron beam when it is deflected by the blanking electrode 3 .

The objective aperture 5 is an aperture that adjusts the opening angle (substantially used for determining the beam diameter of the electron beam) when the objective lens 8 narrows the electron beam and irradiates it onto the sample 11 .

The electron detector 6 detects secondary electrons and reflected electrons emitted when the sample 11 is planarly scanned while irradiating the sample 11 with an electron beam narrowed down, and is a semiconductor device such as an MCP, a scintillator, or an APD.

The deflection electrode 7 scans the surface of the sample 11 with a finely focused electron beam, and is usually a two-stage deflection system (X-axis, Y-axis).

The objective lens 8 narrows the electron beam and irradiates it onto the sample 11 . In this case, two stages are provided: the illustrated differential exhaust aperture A1 below the objective lens 8 and the differential exhaust aperture A2 immediately above it.

The differential pumping aperture 9 is an aperture having a small hole provided between the lens barrel and the vacuum chamber 10, and is used for two-stage differential pumping. Here, an acceleration voltage is applied between the lowest stage differential pumping aperture A1 and the sample 11, and secondary electrons emitted when the sample 11 is irradiated with a narrowly focused electron beam are accelerated upward and deflected so as to pass through a small hole of the differential pumping aperture A1 (described later with reference to FIG. 2).

Here, a plurality of stages (two stages in FIG. 1) of differential pumping apertures 9 are used to block the low vacuum of the vacuum chamber 10 so as not to affect the ultra-high vacuum of the electron gun 1 that generates the electron beam. Normally, one aperture can create a pressure difference of about three digits.

A vacuum chamber 10 is a vacuum container for storing a sample 11 and the like, and is evacuated by a TMP (turbo-molecular pump) 14 or a dry pump (not shown) for roughing.

A sample 11 is a sample such as a wafer or a photomask.

The XY stage 12 carries the sample 11 and moves it in the X and Y directions. Movement is performed in conjunction with a laser interferometer (not shown) for precise movement.

The holder 121 fixes the sample 11 to the XY stage 12, electrically insulates the sample 11 from the XY stage 12, and applies an acceleration voltage to the sample 11. It is desirable that the holder 121 has such a shape as to always form a uniform electric field on the surface of the sample 11 when the XY stage 12 is moved. Especially in the peripheral area, distortion of the electric field is likely to occur, so a structure that avoids such distortion is desirable.

A high voltage power supply 21 generates a high voltage to be supplied to the electron gun 1 .

The electron beam control circuit 22 emits an electron beam from the electron gun 1, converges it with the condenser lens 2, narrows it down with the objective lens 8, irradiates it on the sample 11, and supplies and controls various voltages such as planar scanning with the deflection electrode 7 in this irradiated state.

The PC 23 is a personal computer, and performs various controls using programs.

The display 24 displays various instruction screens, images, and the like.

Gas 25 is an ionizable gas (eg, N2, O2, etc.).

A vacuum controller 26 controls the evacuation of the vacuum chamber 10, the electron gun 1, and the like.

The mass flow 27 reduces the pressure of the gas 25 to a predetermined pressure, controls the flow rate to a predetermined flow rate, and causes the gas to flow into the vacuum chamber 10 .

A pressure sensor 28 measures the pressure in the vacuum chamber 10 .

The stage controller 29 precisely controls the movement of the XY stage 12 to a designated position. Movement is precisely controlled based on a laser interferometer (not shown).

Next, the operation of the configuration of FIG. 1 will be described in detail.

(1) Conventionally, it was thought that applying a bias voltage to the sample 11 in a low vacuum would cause discharge, so application of a bias voltage has been avoided. However, when the present inventor conducted an experiment, it was found that the battery does not discharge at a voltage lower than 400 V, which is completely different from conventional wisdom (see FIG. 5 described later).

(2) Based on this finding, a sample voltage applying device 13 or the like was newly provided for applying a bias voltage or a potential difference so that secondary electrons or reflected electrons generated on the surface of the sample 11 are accelerated in the direction perpendicular to the sample 11.

The degree of vacuum of the vacuum chamber 10 is preferably in the range of about 0.1 to 100 Pa, which is realized by introducing oxygen or nitrogen gas into the vacuum chamber 10 with the mass flow controller 27 . The energy of the primary electron beam to be used varies depending on the measurement object and purpose, and ranges from 0 to 15 KV. The current ranges from several pA to several 100 nA.

Voltage application to the sample 11 is performed by applying a required voltage between the holder 121 holding the sample 11 and the differential exhaust aperture 9 . Since the sample holder 121 is made of metal and covers the back surface of the sample 11, it functions as a conductor flat plate. For a sample 11 having conductivity as a whole, such as a wafer, a voltage may be applied directly to a portion of the sample 11 exhibiting conductivity using a needle or the like. It is also effective to dispose an electrode forming a large capacitance on the rear surface of the sample 11 with an insulator interposed therebetween.

(3) It is not always necessary to supply a current to the sample 11 because it is sufficient to generate a potential difference such that the secondary electrons emitted from the surface of the sample 11 are directed toward the electron detector 6 .

The voltage applied to the sample 11 is a constant voltage in the range of about 0V to 400V which does not cause discharge. If the surface of the structure in the vacuum chamber 10 is made as smooth as possible so that an abnormal concentrated electric field does not occur at a specific location, it is possible to prevent discharge when a higher voltage is applied.

(4) When actually used, it is desirable to apply a discharge voltage of 400 V or more in advance to deliberately cause discharge to remove projections that cause abnormal concentrated electric fields, and to apply aging. It is desirable to install a safety device such as an interlock so as not to apply a voltage of 400 V or higher, which would cause discharge, so that discharge does not occur during operation of the device.

If necessary, the bias voltage may be applied only at the time of measurement, or the voltage application may be controlled in synchronization with certain timing of measurement or electron beam scanning. Alternately, an alternating current of a certain frequency may be applied.

FIG. 2 shows an explanatory diagram (part 1) of the effect of the substrate bias of the present invention.

FIG. 2(a) shows a diagram for explaining the effect of the conventional device without bias, and FIG. 2(b) shows a diagram for explaining the effect of the present invention with bias.

FIG. 2(a) shows how secondary electrons are emitted when no bias is applied to the sample, and FIG. 2(b) shows how secondary electrons are emitted when a bias is applied to the sample.

2(a) and 2(b), the sample 11 is a target sample (such as a wafer or mask) to be irradiated with the primary electron beam narrowed by the objective lens 8 of FIG.

The differential pumping aperture 9 here has a two-stage structure consisting of a lower differential pumping aperture A1 and an upper differential pumping aperture A2 of the differential pumping aperture 9 in FIG. Positive bias voltages V1 and V2 (positive voltages for accelerating secondary electrons) are applied to the differential exhaust apertures A1 and A2, respectively.

In distribution (1), the secondary electrons emitted from the sample 11 in FIG. 2(a) spread in the radial direction (Gaussian distribution) as shown. In this state, radiation is emitted with the same distribution as with the bias of the present invention in FIG. 2(b).

In distribution (2), no bias voltage (positive acceleration voltage) is applied to the differential pumping aperture A1 in (a) of FIG. 2, and the potential is 0. Therefore, the secondary electrons emitted from the sample 11 in all directions repeatedly collide with the gas in the low vacuum in the vacuum chamber 10 and are scattered, resulting in a secondary electron distribution having a wide opening angle distribution as shown in the figure. As a result, the proportion of secondary electrons passing through the small hole in the center of the differential pumping aperture A1 is reduced (2/3 are blocked in some cases), and only a portion of the secondary electrons pass through (see FIG. 5).

On the other hand, in the case of FIG. 2(b) with the bias of the present invention, a positive bias voltage (+400 V or less) is applied to the differential exhaust aperture A1, and the secondary electrons are accelerated upward, so that the distribution of secondary electrons hardly spreads and has a small angular distribution as shown. Therefore, most of the secondary electrons can pass through the small hole in the center of the differential exhaust aperture A1.

In distributions (3) and (4), no bias voltage is applied to the differential exhaust aperture A2 in (a) of FIG.

On the other hand, in FIG. 2(b), a positive bias voltage is applied to the differential exhaust aperture A2 and the secondary electrons are accelerated upward, so the distribution of secondary electrons hardly spreads and has a small angular distribution as shown. Therefore, most secondary electrons can pass through the small hole in the center of the differential pumping aperture A2.

As described above, when no bias is applied, the secondary electrons generated on the surface of the sample 11 have very low energy and fly out in all directions at low velocities. Among the low-velocity electrons emitted from the sample 11 in all directions, the secondary electrons whose emission angle is away from the axis travel outward from the axis without being affected by the magnetic field of the objective lens. There are many parts that do not reach the vessel 6 (described in detail using FIG. 5).

On the other hand, when a bias is applied between the differential exhaust apertures A1 and A2 and the sample 11, the secondary electrons are accelerated in the direction perpendicular to the sample 11, so they travel along the trajectory. Since electron beam scanning is performed over a range of tens of microns at most, electrons rise vertically from that range. Furthermore, when the accelerated secondary electrons enter the magnetic field of the objective lens 8, they strongly interact with the magnetic field, are narrowed in a spiral at the cyclotron frequency, and rise from the surface of the sample 11 like a tornado. As a result, the electrons are directed toward the electron detector 6 and detected without colliding with the inner wall of the vacuum chamber 10 or the outer peripheral portions of the holes of the differential pumping apertures A1 and A2. This effect keeps the aperture clean and free of contamination. If the aperture is contaminated, it will be charged and the electron beam will be bent, but this does not occur in the present invention.

Here, when a bias voltage is applied to the sample 11, as a secondary effect, the ions are strongly attracted to the surface of the sample 11, and the number of ions floating in the space through which the primary electron beam passes or the space directly above the point where the primary electron beam is irradiated is reduced. Due to this effect, the antistatic effect on the surface of the sample 11 becomes stronger, primary electron scattering by ions is reduced, and the spot diameter of the primary electron beam can be kept small.

FIG. 3 shows an explanatory diagram (part 2) of the effect of the substrate bias of the present invention.

FIG. 3(a) shows a diagram for explaining the effect of the conventional device without bias, and FIG. 3(b) shows a diagram for explaining the effect of the present invention with bias.

FIG. 3(a) shows how secondary electrons are emitted when no bias is applied to the sample, and FIG. 3(b) shows how secondary electrons are emitted when a bias is applied to the sample.

In FIGS. 3A and 3B, the sample 11 is a target sample (such as a wafer or mask) to be irradiated with the primary electron beam narrowed by the objective lens 8 of FIG.

The differential pumping aperture 9 here has a two-stage structure consisting of a lower differential pumping aperture A1 and an upper differential pumping aperture A2 of the differential pumping aperture 9 in FIG. Positive bias voltages (100 to 200 V) of V1 and V2 are applied to the differential exhaust apertures A1 and A2, respectively.

In FIG. 3(a), the secondary electrons emitted from the sample 11 repeatedly collide with the gas in the low vacuum of the vacuum chamber 10 as already described in FIG. 2(a) and are scattered in all directions.

On the other hand, in the case of FIG. 3(b) with the bias of the present invention, since a positive bias voltage of 100 to 200 V is applied to A1 on the lower side of the differential pumping aperture 9, the secondary electrons are accelerated upward and almost pass through the small hole in the center of the differential pumping aperture A1. Furthermore, since a bias voltage of 100 to 200 V is applied to the differential pumping aperture A2, the electrons are further accelerated and almost pass through the small hole in the center of the differential pumping aperture A2, and are detected by the electron detector 6.

As described above, in FIG. 3B, by applying a bias voltage (positive acceleration voltage) between the sample 11 and the differential pumping aperture 9, the secondary electrons emitted from the sample 11 are accelerated upward to pass through the small hole at the center of the differential pumping aperture A1, travel upward while spiraling on the axis of the magnetic field of the objective lens 8, and then accelerated upward by the bias voltage of the differential pumping aperture A2 to pass through the small hole at the center of the differential pumping aperture A2. Almost all secondary electrons emitted from the sample 11 can be detected and amplified by the detector 6 .

FIG. 4 shows an explanatory diagram of substrate bias voltage and image SNR of the present invention. Here, FIG. 4 shows changes in the image SNR when a bias voltage is applied between the sample 11 and the differential exhaust aperture 9, the horizontal axis represents the substrate bias voltage (V), and the vertical axis represents the image SNR. The image SNR used here is obtained by dividing the average luminance of all the pixels forming the image by the luminance variance. The experiment was conducted with a primary electron beam energy of 1.5 KV and 10 pA and a vacuum chamber vacuum degree of 30 Pa.

In FIG. 4, when the bias voltage is gradually applied from 0, the image becomes brighter and brighter and the luminance dispersion becomes smaller. This indicates that the directions of the secondary electrons are aligned and a large amount of the secondary electrons pass through the differential pumping aperture 9 and reach the electron detection device 6 . As a result, the image SNR is nearly doubled at a bias of 200V. Thus, bias application is highly effective in improving the image SNR. Above 200 V, saturation is observed, and above 400 V, discharge occurs, so a voltage below that is desirable (see FIG. 5).

FIG. 5 shows an illustration of the vacuum chamber pressure and discharge voltage of the present invention. The horizontal axis represents vacuum chamber pressure (Pa), the right end represents atmospheric pressure, the left end represents high vacuum, and the vertical axis represents minimum discharge voltage (V).

In FIG. 5, when the voltage (bias voltage) applied between the sample 11 in FIGS. 1 and 3 and the differential pumping aperture 9 was measured, the minimum discharge voltage was measured when the pressure in the vacuum chamber 10 was changed.

Therefore, as already described with reference to FIG. 4, if the substrate bias voltage (V) is maintained at about 200 V or less, no discharge occurs even if the vacuum chamber 10 is at a low vacuum, and the secondary electrons emitted from the sample 11 are accelerated toward the differential pumping aperture 9 and passed through the small hole at the center of the differential pumping aperture A1. When the SNR is doubled, the throughput can be improved proportionally and the inspection speed can be more than doubled.

FIG. 6 shows a block diagram of another embodiment of the present invention. This FIG. 6 shows an example in which an ion generation source is provided separately.

Here, the reason why the ion generation source is separately provided is that the scanning electron microscope of FIG. That is, the electron beam for observation and the electron beam for ion generation are shared. Therefore, the amount of the primary electron beam and the amount of generated ions are in a proportional relationship and are constrained to each other. The primary electron beam dose is usually optimized for observation, and ion generation is subordinate. Since the amount of ions that can be generated with respect to a unit amount of primary electron beam (amount of current) is proportional to the amount of gas present in the vacuum, the only way to increase the amount of ions is to introduce gas and further lower the degree of vacuum. Reducing the degree of vacuum will degrade the image SNR, making it impossible to perform the necessary measurements.

Therefore, FIG. 6 is characterized in that it has an electron beam type ion generator, which has an electron beam generator for generating ions separately from the primary electron beam for observation, in order to eliminate the restrictions of FIG. 1 described above.

In FIG. 6, 1 to 13 are the same as 1 to 13 in FIG. 1, so the description is omitted.

The electron gun 31 is a simple device that generates a dedicated electron beam for ionization.

The electron beam 32 is an electron beam emitted from the electron gun 31 and is used to irradiate and ionize the gas (gas) in the low vacuum in the vacuum chamber 10 .

The ion acceleration voltage 33 is a voltage that accelerates the ions generated by the electron beam 32 toward the sample 11. Here, as shown in the figure, the voltage is applied to an electrode (such as a disk electrode with a hole in the center or a flat plate mesh electrode) that allows the ions to reach the irradiation area of the sample 11 with the primary electron beam.

The Faraday cup 34 collects the remaining electron beams that are not used for ionization when the electron beams 32 collide with the gas in the vacuum chamber 10 (once they are inside, they do not come out and do not become a noise source).

The ion flow 35 is an ion flow in which ions generated by the electron beam 32 are accelerated by the ion acceleration voltage 33 and flow toward the primary electron beam irradiation region of the sample 11 .

The gas 25 is an ionizable gas (gas) such as N2 or O2.

The gas flow 27 controls the supply of gas at a constant flow rate (or constant pressure, etc.) to the vacuum chamber 10 according to the pressure inside the vacuum chamber 10 detected by the pressure sensor 28 .

The pressure sensor 28 detects the pressure inside the vacuum chamber 10 .

The operation of the configuration shown in FIG. 6 will be described in detail below.

(1) The electron gun 31 dedicated to ionization shown in FIG. 6 only needs to emit a required amount of electron beam having the energy required for ionization. Therefore, unlike the electron gun 1 shown in FIG. 1, a large-scale lens and control device are not required.

(2) The electron beam emitted from the dedicated electron gun 31 collides with the gas in the vacuum chamber 10 to be ionized, and then 100% absorbed by a Faraday cup or the like so as not to affect image formation.

(3) Since the electron beam generated by the electron gun 31 is as large as 100 μA or more, a large amount of ionization occurs when the electron beam is applied to gas in a low-vacuum state.

(4) Since ionization of gas by electron beam collision is stochastic, as long as non-ionized gas molecules exist in the space through which the electron beam passes, the amount of ion gas increases in proportion to the amount of irradiation current. However, if all the gas present in the space is ionized, saturation will occur and the amount of ions will not exceed that level. If it is generated in too large a quantity, problems will occur. Alternatively, the amount of gas in the vacuum chamber 10 is reduced to adjust the amount of generated ions.

As described above, an electron beam is introduced into the vacuum chamber 10 from the dedicated electron gun 31, the gas in the vacuum chamber 10 is ionized, and the primary electron beam irradiation region of the sample 11 is irradiated with the ionized gas to flow, thereby neutralizing the charges generated by the primary electron beam irradiation of the sample 11.

FIG. 7 shows a block diagram of another embodiment of the present invention. This FIG. 7 is characterized in that an ion generator using a vacuum ultraviolet LED is arranged in the vacuum chamber 10 .

A detailed description will be given below.

(1) Recently, an ultra-compact DUV-LED has been developed and put into practical use by Japanese Nobel Prize-winning technology, so it has become possible to easily generate vacuum ultraviolet rays with a wavelength of the order of 200 nm (about 5 eV) in the vicinity of the sample 11 placed in the vacuum chamber 10. However, since the energy required to ionize the gas is 10 eV or more, nitrogen and oxygen cannot be ionized even if the gas is directly irradiated with ultraviolet light generated by the LED.

(2) Therefore, as shown in FIG. 6, in the present invention, ions are generated in two steps.

(2-1) In the first step, vacuum ultraviolet rays generated by the UV-LED are used to generate a large amount of electrons by the photoelectric effect. A large amount of electrons are generated by exposing a metal or compound (titanium, aluminum, iron, Ni or GaAs, barium oxide, LaB6, etc.) having a work function of 5 eV or less to the vacuum ultraviolet rays.

(2-2) In the second stage, the generated electrons are used to ionize the gas. Specifically, by accelerating electrons generated using vacuum ultraviolet rays from several hundred volts to several kilovolts, the electrons have the energy necessary to ionize the gas, and collide with the gas present in the vacuum chamber 10 to generate ions.

(3) On the other hand, when the vacuum ultraviolet rays generated by the UV-LED strike the surface of the sample 11, unwanted secondary electrons are emitted from the surface of the sample 11 due to the photoelectric effect and become noise, or the conductivity of the surface of the sample 11 changes or the sample 11 deteriorates, causing a change in the state of the sample 11 and causing problems in measurement. .

(4) Since the amount of electron beam current used for measurement with a scanning electron microscope or the like is very small, on the order of pA to nA, even if the efficiency of converting DUV-LED light into electrons is considerably low, it is possible to generate the necessary amount of electrons to generate gas ions for antistatic purposes. For example, the output of currently developed and commercially available DUV-LEDs is about 10 mW. Therefore, even if the conversion efficiency to secondary electrons generated by irradiating a secondary electron generating material such as a metal with the light is only 1%, it is possible to take out an electron emission current of about 1 μA, and if this is accelerated and irradiated to a gas, more than enough ions are generated. This means that, for example, if the amount of electrons ten times that of the primary electron beam is used to generate ions, the same amount of ions can be generated even if the degree of vacuum in the vacuum chamber 10 is one digit higher. It is easy to see that a 10-fold increase in vacuum results in a higher image SNR.

FIG. 8 shows an example of the ion generator of the present invention.

FIG. 8(a) shows an ion generator that generates positive ions, and FIG. 8(b) shows an ion generator that generates negative ions. A detailed description will be given below.

(1) If the amount of ions is too large, even if a sufficient antistatic effect is obtained, the ions may interact with primary electrons and secondary electrons, leading to deterioration in image SNR. In other words, it is necessary to adjust the amount of ions on the surface of the sample 11 so that they are necessary and sufficient. Since the amount of charge on the surface of the sample 11 is in a proportional relationship with the amount of the irradiated electron beam, it is desirable that the amount of ions to be generated is in a proportional relationship with the amount of the electron beam used by the scanning electron microscope or the like.

(2) Therefore, the amount of electron beam is measured and controlled so that the amount of primary electron beam and the amount of generated ions are proportional. The amount of generated ions may be changed by changing the luminous power of the UV-generating LED (DUV-LED 51 in FIG. 8) or by changing the acceleration voltage (electron acceleration voltage 54 in FIG. 8) applied to the generated electrons. Of course, the gas introduction amount may be changed.

(3) The generated ions are transported to the vicinity of the electron beam irradiation area on the surface of the sample 11, and the antistatic effect is produced only after that. In order to transport the ions generated by the electron beam irradiation to the electron beam irradiation area, an electric field is applied to the ion generator using a mesh-shaped ion acceleration electrode (ion acceleration electrode 55 in FIG. 8) to accelerate the ions in one direction and create an ion flow.

(4) In general, the charging of the sample 11 occurs both positively and negatively, but the polarity may differ depending on the location of the sample 11 (when positive and negative charging regions exist simultaneously on one sample).

(5) Therefore, three modes 1, 2, and 3 shown in (a), (b), and (c) of FIG. 9, which will be described later, are realized.

Mode 1: Apply an electric field so that positive ions are carried to the sample surface (see FIG. 8(a)).

Mode 2: Apply an electric field so that negative ions are carried to the sample surface (see FIG. 8(b)).

Mode 3: In the case of partial electrification with different polarities, an alternating electric field is applied so that both positive and negative ions are carried to the sample surface (switching between (a) and (b) in FIG. 8 alternately). Specifically, the polarity of the ion flow control circuit 56 is alternately switched between positive and negative so that positive ions and negative ions alternately reach the vicinity of the electron beam irradiation area. By changing the ratio of positive and negative ions, the ratio of positive and negative ions can be freely controlled. Pulse control may be performed. The ion flow rate can also be changed by changing the applied voltage.

By doing so, it is possible to provide an appropriate amount of ions with an appropriate polarity according to the charging state of the surface of the sample 11 irradiated with the electron beam.

(6) The gas supplied to the ion source may be a gas normally present in the vacuum chamber or a gas source such as a gas cylinder introduced into the vacuum chamber. When the gas existing in the vacuum of the vacuum chamber is used as the ion gas source, the generated ions do not flow in a specific direction, so the magnitude and direction of the ion flow can be freely controlled by applying a bias voltage to create an appropriate ion flow as described above. On the other hand, when directly connected to the gas supply source, the ion flow flows in one direction due to the pressure difference, so a bias power supply and electrodes for creating the ion flow are not required, but the ion flow cannot be freely controlled. Both can be used properly according to the purpose.

(7) Furthermore, in order to efficiently guide the ions emitted from the ion generator shown in FIG. 8 to the sample surface, a voltage may be applied between the sample 11 and the ion generator 41 using the electric field controller 42 shown in FIG.

(8) If ions can be generated separately from the primary electron beam as shown in FIGS. 6 and 7, the efficiency of ion generation can be increased, so that a large amount of ions can be generated without lowering the degree of vacuum in the vacuum chamber 10. As a result, the interaction between the primary electrons and the gas in the vacuum chamber 10 is reduced, the effective amount of irradiation current is increased, the SNR of the image is improved, the beam spot size can be kept small, and the resolution of the scanning electron microscope can be improved.

(9) The amount of ions required to remove charges from the surface of the sample 11 is the same whether the primary electron beam is used or when ions are separately generated as in this embodiment. Therefore, the secondary electrons generated on the surface of the sample 11 by the primary electron beam have the same probability of colliding with the ions before reaching the electron detector 6. Therefore, simply generating the ions with another device does not improve the image SNR much. However, by applying the bias voltage of the present invention between the sample 11 and the differential pumping aperture 9 to capture many secondary electrons in the direction of the electron detector 6, even if the secondary electrons generated on the surface of the sample 11 collide with ions and are scattered, most of them are detected by the electron detector 6, so that the image SNR can be improved.

The configurations shown in FIGS. 8 and 9 will be briefly described below.

FIG. 8(a) shows an example of an ion generator 41 that generates positive ions, and FIG. 8(b) shows an example of an ion generator 41 that generates negative ions.

8A and 8B, the DUV-LED 51 is a semiconductor element that generates ultraviolet rays (vacuum violet ultraviolet rays, 200 nm (approximately 5 eV)), and emits the generated ultraviolet rays to an electron generating material 52 to generate electrons 53.

The electron-generating material 52 generates electrons 53 by ultraviolet rays emitted from the DUV-LED 51 .

The electrons 53 are electrons emitted when the electron generating material 52 is irradiated with ultraviolet rays.

The electron acceleration voltage 54 accelerates the electrons emitted from the electron generating material 52, collides with the gas (N2, O2, etc.) existing in the space between the electron generating material 52 and the DUV-LED 51, and ionizes (ionizes into positive ions and negative ions).

The ion acceleration electrode 55 accelerates either positive ions or negative ions ionized by collision of electrons accelerated by applying the electron acceleration voltage 54 to the gas, and emits an accelerated ion flow (positive) (FIG. 8(a)) or an accelerated ion flow (negative) (FIG. 8(b)) to the outside on the right side. When a positive voltage is applied to the ion acceleration electrode 55, the accelerated ion current (positive) shown in FIG. 8A is emitted, and when a negative voltage is applied to the ion acceleration electrode 55, the accelerated ion current (negative) shown in FIG. 8B is emitted. By switching the voltage applied to the ion acceleration electrode 55 between positive and negative, it is possible to switch between positive and negative ions for emission.

FIG. 9 shows an ion mode illustration of the present invention.

FIG. 9(a) represents mode 1. FIG. In this mode 1, a positive voltage pulse is applied to the ion acceleration electrode 55 as shown in FIG. 8A to emit a pulsed positive ion current.

(b) of FIG. 9 represents mode 2 . In this mode 2, as shown in FIG. 8B, a negative voltage pulse is applied to the ion acceleration electrode 55 to emit a pulsed negative ion current.

(c) of FIG. 9 represents mode 3. FIG. In this mode 3, mode 1 and mode 2 are alternately switched in a pulse form, and the voltage of the ion flow control circuit 56 is alternately switched between positive and negative as shown in FIGS.

1 is a configuration diagram of an embodiment of the present invention; FIG. FIG. 3 is an explanatory diagram (part 1) of the effect of the substrate bias of the present invention; FIG. 2 is an explanatory diagram (part 2) of the effect of the substrate bias of the present invention; FIG. 4 is an explanatory diagram of substrate bias voltage and image SNR according to the present invention; It is an explanatory view of vacuum chamber pressure and discharge voltage of the present invention. FIG. 4 is a configuration diagram of another embodiment of the present invention; FIG. 4 is a configuration diagram of another embodiment of the present invention; It is an example of an ion generator of the present invention. It is ion mode explanatory drawing of this invention.

1: electron gun 2: condenser lens 3: blanking electrode 4: blanking aperture 5: objective aperture 6: electron detector 7: deflection electrode 8: objective lens 9: differential exhaust aperture 10: vacuum chamber 11: sample 12: XY stage 121: holder 21: high voltage power supply 22: electron beam control circuit 23: PC (personal computer)
24: Display 25: Gas 26: Vacuum controller 27: Mass flow 28: Pressure sensor 29: Stage controller 31: Electron gun 32: Electron beam 33: Ion acceleration voltage 34: Faraday cup 35: Ion flow 41: Ion generator 42: Electric field controller 43: Ion flow 51: DUV-LED
52: Electron generating material 53: Electron 54: Electron acceleration voltage 55: Ion acceleration electrode 56: Ion flow control circuit A1, A2: Differential exhaust aperture

Claims (3)

In a scanning electron microscope provided with a secondary electron detector for detecting emitted secondary electrons by scanning a plane while irradiating a sample with a narrowly focused electron beam,
a small aperture through which the electron beam passes between an objective lens for narrowing the electron beam and the sample movably held in the sample chamber;
a second electron beam generated separately from the electron beam collides with a gas in a sample chamber to ionize the second electron beam into positive and negative ions; a trapping means is provided for trapping the second electron beam that does not contribute to the ionization; A sample voltage applying device is provided between the sample and the aperture for applying a predetermined bias voltage of 400 V or less in a range that does not discharge,
The sample voltage applying device applies to the aperture a voltage that accelerates secondary electrons emitted from the sample toward the aperture, deflects the secondary electrons emitted from the sample to pass through a small hole in the aperture, detects the passing secondary electrons with the secondary electron detector to increase the secondary electron detection ratio, applies the bias voltage within a range in which no discharge occurs to suppress discharge, and neutralizes negative and positive charges due to the irradiation of the electron beam to the sample to neutralize the charge. 2. A scanning electron microscope as claimed in claim 1, characterized in that the suppression and background noise are reduced to improve the S/N. In a secondary electron detection method for a scanning electron microscope provided with a secondary electron detector for detecting secondary electrons emitted by scanning a sample while irradiating a sample with a narrowly focused electron beam,
a small aperture through which the electron beam passes between an objective lens for narrowing the electron beam and the sample movably held in the sample chamber;
an ion generator for irradiating a surface of a sample in the sample chamber with the ionized positive and negative ions in an amount sufficient to neutralize the negative and positive charges generated by the irradiation of the electron beam;
A secondary electron detection method for a scanning electron microscope, wherein the ion generator irradiates the sample with positive and negative ions to prevent charging.

Images