NIST VUV Metrology Programs to Support Space-Based ResearchRobert E. Vest, Yaniv Barad, Mitchell L. Furst, Steve Grantham, Charles TarrioElectron & Optical Physics Division, NIST, Gaithersburg, MD, 20899

and

Ping-Shine ShawOptical Technology Division, NIST, Gaithersburg, MD, 20899

Abstract

Vacuum ultraviolet (VUV) radiation, spanning the electromagnetic spectrum from about 2 nm (620 eV) to 200 nm (6.2 eV) has long been important in astronomy, solar physics, and Earth observing systems, among other applications. The National Institute of Standards and Technology (NIST) has several programs to serve the VUV user community, from the Synchrotron Ultraviolet Radiation Facility (SURF III) – a standard of irradiance from 2 nm to 400 nm – to measurement and calibration services for mirrors, photodiodes, and filters. We have recently reduced the uncertainty of our extreme ultraviolet (EUV) detector calibrations by implementing an absolute cryogenic radiometer on one of the SURF beamlines, and have effected several improvements to the EUV detector calibration beamline at SURF. We continue to investigate wide-bandgap semiconductors for use as solar-blind detector technologies, and have recently obtained quantum efficiency and uniformity data from 1 cm2 active area GaN and SiC photodiodes.

1. Introduction

The National Institute of Standards and Technology (NIST) has a wide variety of programs for the calibration of instruments and components for space-based research in the vacuum ultraviolet. Many of these programs have been in existence since the 1960s, and have provided calibration support to NASA and international missions since that time.

Radiometric calibration of detectors, sensor packages, and entire spacecraft instruments is available in the vacuum-ultraviolet (VUV) spectral range from 2 nm to 254 nm. NIST issues transfer standard detectors, either solar-blind photoemissive detectors or solid-state Si photodiodes, in this spectral range with relative expanded uncertainties from 2 % to 20 %. (All uncertainties in this paper are stated as expanded uncertainty with coverage factor k = 2, i.e., 2σ uncertainties.) End-to-end calibration of entire instruments is available at several synchrotron beamlines using either detector-based or source-based standards. NIST is collaborating with several developers of novel detector technologies, especially in the characterization of new wide-bandgap semiconductors that can be used as solar-blind, solid-state photodiodes.

In addition to radiometric calibrations, NIST has facilities for the measurement of reflectivity as a function of position, angle of incidence, and wavelength. Many

multilayer mirrors for extreme ultraviolet (EUV) space-based telescopes have been calibrated here. Facilities are available for the measurement of the transmission of thin-film and crystalline windows throughout the VUV spectral region. The zero-order and diffraction efficiency of EUV transmission gratings can be measured as well.

2. Sources and Calibration Facilities

NIST operates a number of facilities for the EUV radiometry program. These include a low-pressure plasma discharge and a normal-incidence monochromator, a laser-produced plasma for pulsed measurements, and a synchrotron radiation facility with several beamlines dedicated to radiometry.

2.1. FUV Detector Calibration Facility

In the Far Ultraviolet (FUV) Detector Calibration Facility a low-pressure plasma discharge source is operated to produce radiation with wavelengths longer than 50 nm. While the source also produces shorter wavelength radiation, a normal-incidence monochromator is used for the measurements here. The source, called a “duoplasmatron,” was originally developed as an ion source, but was adopted as a source of neutral and singly ionized atomic and molecular emission (Samson, 2000). The plasma is generated by electrons emitted from a hot filament and accelerated through a gas by an applied electric field. Permanent magnets surround the plasma chamber and generate an axial magnetic field that pinches the plasma and increases the source’s brightness.

This facility has a 1 m focal-length, normal-incidence, spherical-grating monochromator to provide radiation to the experimental chamber. The duoplasmatron is operated with one of several gases (He, Kr, Ar, Ne, or H2) to produce the desired spectrum. All of these source gases produce atomic (primarily neutral and singly ionized) emission lines. The H2 plasma also emits many molecular lines and a molecular continuum at wavelengths longer than 170 nm.

Instrumentation exists for a variety of experiments in this facility. The most common measurement is detector quantum efficiency, however the capability to measure the transmission of thin-film filters and crystalline windows and the spatial uniformity of a photodiode’s responsivity is also available. The detector calibration can be repeated during an extended exposure to determine the detector’s radiation hardness. The typical wavelength range of this facility is 50 nm to 254 nm. Although longer wavelengths are available from the source and monochromator, another laboratory at NIST is better suited to perform calibrations in the near ultraviolet and visible spectral ranges (NIST, 2004).

All three of the transfer standard detector types described in Section 3 are calibrated in this facility. The absolute detector that is the head of the calibration chain is either a double-plate ionization chamber (described in Section 4.2) from 50 nm to 92 nm or an absolute cryogenic radiometer (described in Section 4.3) from 130 nm to 254 nm. The ionization chamber is operated in this facility, but the radiometer is operated on Beamline-4 (BL-4) at the Synchrotron Ultraviolet Radiation Facility (SURF III) described in Section 2.3.3.

2.2. Pulsed Radiometry for EUV Lithography

In the EUV Lithography (EUVL) Pulsed Radiometry Facility we operate an EUV laser-produced plasma source (Parra et al., 2002) that is similar to sources being considered for use in commercial lithography systems. The NIST source injects gas clusters or droplets into a vacuum chamber by means of a pulsed valve. A Qswitched laser is focused onto the gas and generates a plasma with high ionization states. The plasma emits in a broad spectrum, but proper selection of the target gas can optimize the source for 13.5 nm, the EUVL operational wavelength. This source can operate with either Kr or Xe as the target gas and uses a 10 W average power Nd:YAG laser at either the 1064 nm fundamental emission or the 532 nm frequency-doubled wavelength. The EUV emission from the plasma is pulsed, with a pulse length of approximately 10 ns. The typical operating frequency is 1 Hz.

This facility is instrumented for the calibration of EUVL source diagnostics tools and EUV detectors in the pulsed illumination conditions found in EUVL wafer exposure systems. Typically, the source is imaged onto both a normalization detector and a second detector, either the detector under test or the in-house working standard detector. The signal from the normalization detector allows us to account for the shot-to-shot variation in the source radiance. The calibration is performed by interchanging the detector under test and the working standard in the vacuum chamber.

All the working standard detectors in this facility are calibrated at the EUV Detector Radiometry Beamline (BL-9) at SURF III (see Section 2.3.4). The calibration is established in a low-power, continuous beam of radiation. We are currently studying (Vest and Grantham, 2003) the response of solid-state photodiodes to short, high-peak-power pulses of radiation to understand how best to transfer low-power, cw calibrations to pulse energies typical of EUVL applications.

2.3. Synchrotron Ultraviolet Radiation Facility (SURF III)

The Synchrotron Ultraviolet Radiation Facility (SURF III) is one of the oldest sources of synchrotron radiation, particularly of EUV radiation, in the world. Currently there are many scientific beamlines with diverse applications that span the spectral range from 2 nm to 400 nm in operation at SURF III. Applications include radiometric calibration of sources and detectors, reflectometry, index of refraction measurements, and accelerated lifetime testing of EUVL multilayer mirrors.

Synchrotron radiation is a continuum emission from the charged particles (typically electrons or positrons) stored in the synchrotron as they are accelerated along a curved trajectory. SURF III is an electron synchrotron. The emission covers a broad spectral range, with a long-wavelength limit in the infrared and a short wavelength limit that is a function of the electron energy and bending radius. The wavelength where the radiated power is maximized is also a function of the electron energy and is typically found in the x-ray portion of the spectrum for electron energies of 1 GeV or more. SURF III is unique among synchrotron radiation facilities in that the maximum electron energy is 400 MeV. The peak of the emission is in the VUV for this energy, making

SURF III an excellent source of synchrotron radiation for VUV calibration programs. Additionally, the emitted spectrum can be shifted by changing the electron energy. In this way, high diffraction orders and scattered light components in a beam can be controlled. Figure 1 shows synchrotron radiation spectra at several electron energies as well as the spectrum from a 3000 K blackbody radiator for comparison.

Synchrotron radiation is the only practical, calculable radiation source available in the VUV. A blackbody, the only other calculable radiation source, would need to be at unrealistic temperatures to emit efficiently at these wavelengths. The radiance of the electron beam, or the irradiance on an aperture, can be calculated from first principles and a few accelerator parameters. To calculate the intensity and spectral distribution of the synchrotron radiation, one must know the number of electrons in the storage ring and two of the following three quantities: electron energy, magnetic field strength, and electron bending radius. This calculability makes SURF III a standard of radiance or irradiance as well as a general continuum source of VUV radiation.

The SURF III synchrotron is a multi-user source, and there are many experiments in progress at any given time. Some of the beamlines most related to metrology for space-based observations are described in the following sections.

2.3.1. NIST/DARPA Reflectometry Beamline (BL-7)

The NIST/Defense Advanced Research Projects Agency (DARPA) reflectometry beamline (BL-7) at SURF III (Grantham et al., 2002; Haass et al., 1994; Tarrio et al., 2003) consists of a set of collecting optics, a varied-line-spaced grating monochromator, and a large reflectometer. The reflectometer is capable of mounting large optics up to 35 cm in diameter and 50 kg in mass. The sample positioning stage affords three degrees of translational freedom and three degrees of rotational freedom. The sample can be placed with an uncertainty of less than 1 mm and pointed with an uncertainty of less than 0.05° about all three axes. A second motion scans a detector in the plane of reflection and perpendicular to the plane of reflection. This second motion allows reflectance measurements to be made on highly curved surfaces where the range of motion in the sample pointing is insufficient to bring the reflected beam into the detector plane.

Reflectivity of normal incidence mirrors, e.g., EUV telescope collectors, is measured with relative uncertainties of less than 1 % in the wavelength range from 11 nm to 17 nm, and with somewhat larger uncertainties at wavelengths down to 5 nm and as long as 35 nm. The reflectometer also can be configured for the measurement of the transmission of thin-film filters and the diffraction efficiency of transmission gratings.

The reflectometry beamline is also used for the radiometric calibration of assembled instruments. The monochromatic beam can be collimated and redirected through an exit flange in the rear of the reflectometer chamber by a set of transfer optics (Tarrio et al., 2005). These optics consist of a plane mirror at the sample position in the reflectometer and a cylindrical mirror that collimates the beam and directs it horizontally through the exit flange. The collimated beam size is approximately 4 mm x 4 mm. With these transfer optics, the beamline can be used to calibrate detectors or whole

instruments. Recently we have operated an absolute cryogenic radiometer (see Section 4.3) as a primary standard detector between 11 nm and 35 nm with this configuration. We have also calibrated two EUVL source diagnostic instruments at 13.5 nm wavelength.

2.3.2. NASA Spectrometer Calibration Beamline (BL-2)

SURF III is used as a standard of irradiance at the NASA Spectrometer Calibration Beamline (BL-2), a white-light beamline that makes use of the calculable nature of synchrotron radiation (Furst et al., 1993; Vest et al., 1999). The photon flux through an instrument’s entrance aperture is calculated from a few accelerator parameters and a calibration is derived by measuring the response of an instrument to the known photon flux. Uncertainties in the flux calculation throughout the 2 nm to 400 nm spectral range are less than 1 %. The electron energy in the SURF III beam can be varied to eliminate or to help characterize high-order diffraction effects in the calibration instrument. The ability to vary the electron energy also allows the calibration of optical systems that lack dispersive elements. For example, silicon photodiodes with integral thin-film filters (Canfield et al., 1994) have recently been calibrated without a monochromator (Woods, 2004).

2.3.3. Normal-Incidence Detector Radiometry Beamline (BL-4)

The Normal Incidence Detector Radiometry Beamline (BL-4) is a high-accuracy general purpose radiometric beamline (Shaw et al., 1998) which covers a spectral range from 130 nm to 600 nm. The beamline consists of a 2 m focal length monochromator, a refocusing system, a test detector chamber, and a liquid-helium cooled cryogenic radiometer (see Section 4.3). The cryogenic radiometer measures the optical power of the monochromatized synchrotron radiation using electrical substitution and serves as the primary scale for the calibration of the power responsivity of test detectors. The uncertainty for power responsivity calibrations in this wavelength range is less than 1 %. This beamline can also be used to measure important radiometric quantities such as irradiance responsivity, material transmittance, and reflectance. We also use the beamline to study the UV radiation damage to solid-state detectors. Recently, a study of the damage to a variety of photodetectors by 157 nm and 193 nm excimer lasers was conducted (Shaw et al., 2005). Characterization of the damage effects also sheds light on the mechanism responsible for the changes in the photodetectors.

2.3.4. EUV Detector Radiometry Beamline (BL-9)

The EUV Detector Radiometry Beamline (BL-9) utilizes the synchrotron radiation from SURF III as a continuum source from 5 nm to 49 nm wavelength (Canfield, 1987; Vest et al., 1999). The incident radiation is monochromatized by a grazing-incidence, toroidal-grating monochromator. The experimental chamber can house up to three sample photodiodes in addition to two calibrated working standard detectors. A single-plate ionization chamber is located upstream of the experimental chamber and can be used as an absolute detector for the calibration of photodiodes.

End-station instrumentation is available for the calibration of detectors and the measurement of filter transmission. It is also possible to mount a small instrument package in the chamber on an x-y translation stage with pitch and yaw rotations. The instrument can then be calibrated as a function of angle of incidence. This has been done for the Solar EUV Monitor on the SOHO spacecraft and a similar rocket-underflight instrument used to track the calibration of the SOHO instrument (see Section 5.3).

3. Transfer Standard Detectors

Since 1970 NIST has issued transfer standard detectors in the VUV. Initially, the program covered only the spectral range from 116 nm to 254 nm. New detector technologies have extended the program’s spectral coverage down to 5 nm. Currently three types of detectors are issued as calibrated transfer standards: a CsTe photoemissive detector with an integrated MgF2 window (116 nm to 254 nm), an Al2O3 windowless photoemissive detector (5 nm to 122 nm), and a windowless radiation-hardened, semiconductive Si photodiode (5 nm to 254 nm). Figure 2 shows the detector types and applicable wavelength ranges.

The general procedure for the calibration of a transfer standard is to compare the response of the transfer standard to that of a working standard detector of the same type. The quantum efficiency of the working standard detector is well known from a series of careful calibrations against one of the absolute detectors operated by NIST. These absolute detectors are described in Section 4. The transfer standard detectors are described in the next sections.

3.1. Windowed CsTe Photodetector

The first VUV windowed transfer standard photodiode was issued in 1970 with a calibration from 116 nm to 254 nm. These devices have evolved over the years, and the present configuration uses a ceramic body with three copper electrodes: an anode, a connection to the photocathode, and a guard ring to prevent leakage currents from appearing in the cathode photocurrent. During the fabrication process, a semi-transparent film of CsTe, the photocathode, is deposited on the MgF2 window and the window is sealed to the ceramic body under ultra-high-vacuum conditions. The ceramic body remains evacuated for the life of the device. Figure 3 shows the physical structure of the NIST windowed transfer standard detector. The typical operating voltage on the anode is +150 V.

Figure 4 shows typical quantum efficiency (detected photoelectrons divided by incident photon number) and responsivity (detected photocurrent divided by incident power) curves for these devices. There can be a great deal of variability in efficiency from one device to another, with the efficiency determined by the stoichiometry of the photocathode and transmittance of the window. The short-wavelength response limit is fixed by the transmission edge of the MgF2 window, and the long-wavelength response limit is typically between 300 nm and 350 nm, with some weak response at longer wavelengths.

3.2 Windowless Al2O3 Photodetector

For over 30 years, NIST has issued windowless transfer standard photodiodes with calibration from 5 nm to 122 nm. The photocathode consists of a quartz substrate with a 150 nm Al film. The native Al2O3 is increased from the natural thickness of approximately 5 nm to 15 nm in an anodizing bath. This thickened oxide provides a stable photocathode that is thin enough to remain linear to photon flux levels of about 1011 s-1 in the middle portion of the calibration range. The photocathode is housed in a polytetrofluoroethylene body with a stainless-steel, cylindrical anode. Figure 5 shows the configuration of this detector. The typical operating voltage on the anode is +60 V to +150 V, depending on wavelength. The photoelectron collection efficiency of the anode will generally be less than unity, so photocurrent must be measured as emission from the photocathode.

Figure 6 shows the typical quantum efficiency and responsivity of a NIST Al2O3 windowless photodiode. The peak efficiency is around 70 nm, but the device responds to radiation throughout the spectral range from 5 nm to 125 nm. The oscillations in the efficiency result from interference effects in the reflectance spectrum of the Al2O3 / Al / SiO2 film system. The quantum efficiency exhibits a minimum at a reflectance maximum.

3.3 Windowless Si Photodiode

The first radiometric-quality Si photodiodes for use in the extreme ultraviolet spectral region were developed in the late 1980s (Canfield et al., 1989). In 1990, NIST began issuing Si devices in the wavelength range from 5 nm to 49 nm. Radiation-induced damage to the SiO2 layer caused these devices to degrade significantly when exposed to radiation in the spectral range in which SiO2 is strongly absorbing. In 1993, radiation-hardened Si photodiodes became available (Korde et al., 1993), and the calibration range was extended to cover the entire VUV spectrum from 5 nm to 254 nm. The Si far ultraviolet transfer standard photodiode is an nonp device with a nitrided SiO2 passivating layer. The nitridation process hardens the oxide against radiation damage. A schematic diagram of the device construction is shown in Fig. 7. Photocurrent should be measured in the anode circuit to prevent any photoemission current from the front surface of the photodiode from appearing in the semiconductive photocurrent (Vest and Canfield, 1997).

Figure 8 shows the quantum efficiency and responsivity of a typical far ultraviolet transfer standard Si photodiode. The drop in responsivity at wavelengths shorter than the Si LII,III edge at 12.4 nm is evidence of a small amount of surface recombination in this device. Si transfer standards with no surface recombination (100 % internal efficiency) have been demonstrated, but some devices with nitrided oxide layers exhibit a small degree of surface recombination.

4. Absolute Detectors

NIST detector calibrations in the VUV trace to one of two types of absolute detector. These detectors do not themselves require calibration because of their operating

principle. The first of these detectors is the ionization chamber. This detector is used both in the single-plate and double-plate mode, which are described in the next sections. The second absolute detector is an absolute cryogenic radiometer (ACR), which is a type of electrical substitution radiometer.

4.1 Single-Plate Ionization Chamber

The single-plate ionization chamber (Samson, 1964) consists of an absorbing volume of gas in an electric field. The gas is ionized by incident photons, and the electric field separates the photoelectrons from the ions, which are collected at the cathode. The electric current measured in the cathode circuit is a direct measure of the number of ions produced. The photon flux, Φ, is determined from the measured ion current, i, the gas density, n, the absorbing path length, L, the photoionization cross-section, σ, of the gas, and the ion yield, γ , per incident photon:

€

Φ =i

qγ(1− e−σnL ), (1)

where q is the electronic charge. By choosing a gas where the incident photons carry less than twice the ionization potential of the gas, the ion yield factor, γ, can be made unity.

Photoelectrons are accelerated by the electric field and can further ionize the gas due to electron-impact ionization. The ions produced by this process are also collected at the cathode, and they cause Eq. (1) to overstate the photon flux. The confounding effect of the electron-impact ionization can be removed by measuring the flux at two or more low gas pressures and making a linear extrapolation to zero gas pressure (Samson and Haddad, 1974). Similarly, a photodetector can be calibrated by comparison to the ionization chamber at multiple gas pressures and the derived quantum efficiency extrapolated to zero pressure.

4.2 Double-Plate Ionization Chamber

The single-plate ionization chamber requires knowledge of several parameters: photoionization cross-section, path length, and gas density (usually determined from pressure and temperature measurements). The uncertainty in the determination of these quantities, particularly the cross-section, contributes to the uncertainty of the flux measurement and to the photodetector calibration. Operating two ionization chambers in series with no gap between them eliminates the need to know these quantities (Samson, 1964). In this double-plate mode, the flux is only a function of the two measured ion currents, i1 from the first plate and i2 from the second plate:

€

Φ =i12

qγ(i1 − i2). (2)

As in the single-plate configuration, the multiple yield factor γ can be made unity by proper selection of the ionization chamber gas. Because of the difference (i1i2) in the

denominator of Eq. (2), the double-plate configuration is only useful when there is a high enough gas pressure to give significantly different ion currents at the first and second ion collector plates. A current ratio, i1/i2, greater than 2.5 has been found to give good results.

4.3 Absolute Cryogenic Radiometer

The most accurate optical power meter available today is the absolute cryogenic radiometer (ACR), which is an electrical substitution radiometer operated at liquid helium temperatures (Shaw et al., 1999). Essentially, the radiometer consists of a receiver cavity made from electro-formed Cu with very thin walls, minimizing the thermal mass. With an upstream optical shutter closed so that no radiation is incident on the cavity, resistive heaters maintain a temperature slightly above the liquid helium point at 4.2 K. The cryogenic temperature reduces the heat capacity of the system and increases the thermal sensitivity. The electrical power required to maintain the temperature is recorded. The optical shutter is then opened to allow incident radiation to be absorbed in the cavity. The absorbed optical power heats the cavity, and the thermal stabilization circuits reduce the electrical power in the heaters by an amount equal to the optical power. The difference in electrical power with the shutter open and closed is the optical power carried by the incident radiation. Figure 9 shows a typical data acquisition sequence. In this way the power is measured directly, and the photon flux is calculated from the power and wavelength.

An ACR has been operated on the Normal-Incidence Detector Radiometry Beamline (BL-4) to establish a radiometric scale in the wavelength range from 130 nm to 254 nm and longer (Shaw et al., 2001). The improved accuracy of the calibrations resulted in a decrease of the stated uncertainty in the calibration of transfer standards from about 10 % to 2 % in this wavelength range. Figure 10 shows the quantum efficiency of a NIST working standard photodiode determined before and after the ACR scale was established.

Recently, an ACR was operated on the reflectometry beamline (BL-7) with the transfer optics (described in Section 2.3.1) installed to establish a new radiometric scale in the spectral range from 11 nm to 35 nm. The ACR has considerably lower uncertainties associated with it than the single-plate ionization chamber used at the EUV Detector Radiometry Beamline (BL-9). Figure 11 shows the quantum efficiency of a photodiode determined before and after the ACR scale was established.

5. Recent Metrology Results

In the following sections, we present some recent calibration activities. These examples are an indication of the measurement capabilities and research programs available to support space-based research and related technology development. NIST has worked with industrial and government collaborators to characterize wide-bandgap photodiodes. We have also worked with university-based researchers to provide calibration support for NASA and international space-based observing platforms.

5.1. Wide-Bandgap Solid-State Detectors

The broadband response of Si photodiodes is the primary difficulty in using these detectors to measure VUV signals in solar physics applications or other environments with a high visible radiation background. The indirect bandgap in Si is only 1.1 eV, corresponding to a long-wavelength cutoff in the detector responsivity of 1.1 µm. Wide-bandgap semiconductor materials have a much shorter cutoff wavelength, corresponding to the higher bandgap energy. Two materials that have received much attention recently are GaN with a bandgap of 3.4 eV and long-wavelength cutoff of 365 nm and SiC with a bandgap in the vicinity of 2 eV to 3 eV and a long wavelength cutoff between 380 nm and 560 nm, depending on the SiC structure.

We have recently measured the quantum efficiency of GaN and SiC photodiodes in 5 mm x 5 mm and 10 mm x 10 mm configurations, as well as the spatial uniformity of responsivity of the 5 mm x 5 mm GaN sample. These samples were designed and fabricated by researchers at Raytheon and the NASA Goddard Space Flight Center (Aslam et al., 2004). Figure 12 shows the quantum efficiency of two 5 mm x 5 mm Schottky devices, one with a continuous Pt film and the other with a Pt mesh patterned onto the active area. The device with the mesh electrode has lower responsivity by a factor equal to the coverage area of the Pt, indicating that electron-hole pairs generated in the exposed GaN are not collected at the Schottky barrier and do not contribute to the photocurrent. This is consistent with the known short minority carrier recombination length in GaN (Chernyak et al., 2000). Figure 13 shows the spatial uniformity of responsivity for a 5 mm x 5 mm GaN. The responsivity varies by about ±4 % over the active area of the device. Figure 14 shows the quantum efficiency of 10 mm x 10 mm GaN and SiC photodiodes.

5.2. EVE Filters

The EUV Variability Experiment (EVE) (LASP, 2004) is one of the instrument suites on NASA’s Solar Dynamics Observatory (SDO) satellite mission (NASA, 2004a). EVE is designed to monitor the solar EUV irradiance with high spectral resolution, accuracy, and time cadence. The EVE grating spectrographs have thin-film filters of various materials to reduce the high diffraction orders present in the spectra. These filters have a 1 mm x 5 mm open aperture and are overfilled by the beam in the EUV Detector Radiometry Beamline (BL-9) at SURF III. We developed a method of measuring the transmission of these filters that relies on measuring the transmission of the overfilled filter and an identical filter mount with no filter installed. The results of the measurements and calculated fits to the data are shown in Fig. 15.

5.3. Solar EUV Monitor

The Solar EUV Monitor (SEM) (McMullin et al., 2002; USC, 2000) is a transmission grating instrument on the NASA/ESA Solar and Heliospheric Observatory (SOHO) spacecraft (Fleck et al., 1995; NASA, 2004b) designed to monitor the solar irradiance in the Al bandpass (17 nm to 50 nm) and in a narrow band (±4 nm) about the 30.4 nm He II emission line. The instrument incorporates an Al filter upstream of a

transmission grating. The SEM channel 2 monitors the zero-order beam for the Al bandpass region and channels 1 and 3 monitor the positive and negative first diffraction orders for the 30.4 nm emission line. The SEM on-board SOHO was calibrated at both the EUV Detector Radiometry Beamline (BL-9) and the NASA Spectrometer Calibration Beamline (BL-2) at SURF II (the predecessor of SURF III) prior to launch. While the SOHO SEM is not retrievable for recalibration, a nominally identical instrument is calibrated at SURF III before and after periodic rocket underflights. These underflights serve to transfer the rocket instrument’s calibration to the spacecraft at the Lagrange point between the Earth and the Sun.

The calibrations at NIST consist of a series of measurements at various angles of incidence to determine the SEM efficiency under a variety of alignment conditions. Figure 16 shows the SEM efficiency as a function of wavelength in all three channels when aligned normal to the incident beam, and Fig. 17 shows the efficiency of one of the 30.4 nm channels at several incident angles. Knowledge of the alignment sensitivity allows the SEM team to correct the data for platform pointing errors.

6. Conclusion

NIST operates several calibration facilities for applications in the VUV spectral range from 2 nm to 400 nm. A low-pressure plasma discharge provides a spectrum of atomic – and in the case of a H2 plasma, molecular – emissions from neutral and singly ionized species from 50 nm to 254 nm. A laser-produced plasma with either Kr or Xe target generates 10 ns pulses of EUV radiation for use in characterizing instruments and detectors for EUV Lithography. SURF III provides a continuum of synchrotron radiation from an electron beam with tunable energy for use in radiometry, reflectometry, and many other applications.

Combining the capabilities of these facilities, NIST can provide measurements of detector quantum efficiency, filter transmission, transmission grating diffraction efficiency, mirror reflectivity, and end-to-end calibration of an entire instrument package at either a source-based or detector-based radiometry facility. These capabilities support a number of past, present, and future space-based research programs.

Figure Captions

Figure 1. The calculated synchrotron radiation spectra emitted from SURF III at various electron energies from 38 MeV to 380 MeV. For comparison, the spectrum from a blackbody at 3000 K is shown. Synchrotron radiation is the only practical, calculable source of radiation in the VUV.

Figure 2. The wavelength coverage of VUV transfer standard detectors. Together, the two photoemissive devices (Al2O3 and CsTe) cover the VUV, as does the solid-state Si photodiode. The Si detector can be calibrated into the infrared portion of the spectrum as well.

Figure 3. Schematic diagram of the CsTe photoemissive detector. Radiation is transmitted through the MgF2 window and causes the CsTe photocathode to emit electrons. An extraction field is generated by biasing the anode.

Figure 4. Quantum efficiency (x, left axis) and responsivity (+, right axis) of the CsTe photoemissive detector.

Figure 5. Schematic diagram of the Al2O3 photoemissive detector. Radiation is incident through the cylindrical anode and strikes the photocathode, an Al film with thickened Al2O3. An extraction field is generated by biasing the anode.

Figure 6. Quantum efficiency (x, left axis) and responsivity (+, right axis) of the Al2O3 photoemissive detector.

Figure 7. Schematic diagram of the Si photodiode. NIST transfer standard Si photodiodes are nonp devices with a radiation-hardened oxide layer. The device is calibrated without bias, i.e., in the photovoltaic mode.

Figure 8. Quantum efficiency (x, logarithmic left axis) and responsivity (+, linear right axis) of the Si photodiode.

Figure 9. Data acquisition sequence from the absolute cryogenic radiometer (ACR). The optical power is measured by determining the difference in the electrical power required to maintain a constant temperature in the receiver cavity with and without radiation incident on the cavity. The measured power plotted here is the electrical power delivered to the heaters.

Figure 10. The far ultraviolet radiometric scale from 125 nm to 254 nm was realized with improved accuracy by operating an ACR on BL-4 at SURF III. The uncertainty margins are the expanded uncertainty (k = 2) of the previous (open) and new (solid) detector quantum efficiency scales. The uncertainty margins for the new scale are about the same size as the plot symbol.

Figure 11. The extreme ultraviolet radiometric scale from 11 nm to 35 nm was realized with improved accuracy by operating an ACR on BL-9 at SURF III. The uncertainty margins are the expanded uncertainty (k = 2) of the previous (open) and new (solid) detector quantum efficiency scales. The uncertainty margins for the new scale are about the same size as the plot symbol.

Figure 12. The quantum efficiency of a 5 mm x 5 mm GaN Schottky photodiode was measured in the VUV. The Schottky metal is Pt, as can be seen from the O3 absorption edge at 51.7 nm. Using a mesh electrode rather than a continuous film decreases the efficiency of the device because the minority carrier diffusion length in GaN is so short that electron-hole pairs generated in the uncoated region are lost to recombination rather than collected at the Schottky junction. Note the long-wavelength limit at 365 nm, corresponding to the GaN bandgap of 3.4 eV.

Figure 13. The spatial uniformity of responsivity for the continuous-film device shown in Fig. 12. The contours are every 2 %, measured relative to the responsivity at the center of the device. The responsivity varies by about ±4 % over the 5 mm x 5 mm active area.

Figure 14. The quantum efficiency of 10 mm x 10 mm wide-bandgap photodiodes made from GaN and SiC.

Figure 15. Measured and calculated transmission values from various filters to be used in the EVE instrument on-board SDO. The measurements were made at BL-9 at SURF III. The calculations are based on the results of fitting the material thicknesses to predict the measured data.

Figure 16. Results of a calibration campaign for the rocket instrument on calibration underflights for the SEM instrument on-board SOHO. The measurements were made at BL-9 at SURF III. Ch. 1 monitors the zero-order transmission of the grating. Ch. 2 and 3 monitor the positive and negative first diffraction orders of the 30.4 nm He II emission.

Figure 17. The calibration sensitivity to alignment of one side channel for 30.4 nm He II emission monitoring in the SEM rocket underflight instrument. The measurements were made at BL-9 at SURF III.

References

Aslam, S., Vest, R. E., Franz, D., Yan, F., and Zhao, Y. (2004), "Large area GaN Schottky photodiode with low leakage current," IEE Electronics Letters, 40, 1080-1081.

Canfield, L. R. (1987), "New Far Uv Detector Calibration Facility at the National Bureau of Standards," Appl. Opt., 26, 3831-3837.

Canfield, L. R., Kerner, J., and Korde, R. (1989), "Stability and Quantum Efficiency Performance of Silicon Photodiode Detectors in the Far Ultraviolet," Appl. Opt., 28, 3940-3943.

Canfield, L. R., Vest, R. E., Woods, T. N., and Korde, R. (1994), "Silicon photodiodes with integrated thin film filters for selective bandpasses in the extreme ultraviolet" in Ultraviolet Technology V, Proc. SPIE 2282 (Eds, Huffman, R. E. and Stergis, C. G.), pp. 31-38.

Chernyak, L., Osinsky, A., Fuflyigin, V., and Schubert, E. F. (2000), "Electron beam-induced increase of electron diffusion length in p-type GaN and AlGaN/GaN superlattices," Appl. Phys. Lett., 77, 875-877.

Fleck, B., Domingo, V., and Poland, A. (1995), "Special issue on Solar and Heliospheric Observatory (SOHO)," Solar Physics, 162, 1-531.

Furst, M. L., Graves, R. M., and Madden, R. P. (1993), "Synchrotron Ultraviolet-Radiation Facility (SURF-II) Radiometric Instrumentation Calibration Facility," Optical Engineering, 32, 2930-2935.

Grantham, S., Tarrio, C., Squires, M. B., and Lucatorto, T. B. (2002), "First results from the improved NIST/DARPA EUV reflectometry facility" in Emerging Lithographic Technologies VI, Proc. SPIE 4688 (Ed, Engelstad, R. L.), pp. 348-353.

Haass, M., Jia, J. J., Callcott, T. A., Ederer, D. L., Miyano, K. E., Watts, R. N., Mueller, D. R., Tarrio, C., and Morikawa, E. (1994), "Variable Groove Spaced Grating Monochromators for Synchrotron Light-Sources," Nucl. Instrum. Methods A, 347, 258-263.

Korde, R., Cable, J. S., and Canfield, L. R. (1993), "One Gigarad Passivating Nitrided Oxides for 100-Percent Internal Quantum Efficiency Silicon Photodiodes," IEEE Trans. Nucl. Sci., 40, 1655-1659.

LASP, "SDO EUV Variability Experiment (EVE)" (Laboratory for Atmospheric and Space Physics, 2004), retrieved Dec 2004, http://lasp.colorado.edu/programs_missions/present/off_site/eve.html.

McMullin, D. R., Judge, D., Hilchenbach, M., Ipavich, F., Bochsler, P., Wurz, P., Bürgi, A., Thompson, W. T., and Newmark, J. S. (2002) "In-flight Comparisons of Solar EUV Irradiance Measurements Provided by the CELIAS/SEM on SOHO" in The Radiomteric Calibration of SOHO(Eds, Pauluhn, A., Huber, M. C. E. and von Steiger, R.) International Space Science Institute, Bern, Switzerland, pp. 135-144.

NASA, "Solar Dynamics Observatory Mission Site" (National Aeronautics and Space Administration, 30 Aug 2004a), retrieved Dec 2004, http://sdo.gsfc.nasa.gov/.

NASA, "Solar and Heliospheric Observatory" (National Aeronautics and Space Administration, 2004b), retrieved Dec 2004, http://sohowww.nascom.nasa.gov/.

NIST, "NIST Optical Technology Division - Photodetector Measurements" (National Institute of Standards and Technology, Aug 2004), retrieved Nov 2004, http://physics.nist.gov/Divisions/Div844/facilities/phdet/phdet.html.

Parra, E., McNaught, S. J., and Milchberg, H. M. (2002), "Characterization of a cryogenic, high-pressure gas jet operated in the droplet regime," Rev. Sci. Instrum., 73, 468-475.

Samson, J. A. R. (1964), "Absolute intensity measurements in the vacuum ultraviolet," J. Opt. Soc. Am., 54, 6-15.

Samson, J. A. R., and Haddad, G. N. (1974), "Absolute photon-flux measurements in the vacuum ultraviolet," J. Opt. Soc. Am., 64, 47-54.

Samson, J. A. R. (2000) Techniques of Vacuum Ultraviolet Spectroscopy, Cruithne Press, Glasgow.

Shaw, P.-S., Larason, T. C., Gupta, R., Brown, S., Vest, R. E., and Lykke, K. R. (2001), "The new UV spectral responsivity scale based on cryogenic radiometry at SURF III," Rev. Sci. Instrum., 72, 2242-2247.

Shaw, P.-S., Gupta, R., and Lykke, K. R. (2005), "Stability of photodiodes under irradiation with a 157-nm pulsed excimer laser," Appl. Opt., 44, 197-207.

Shaw, P. S., Lykke, K. R., Gupta, R., O'Brian, T. R., Arp, U., White, H. H., Lucatorto, T. B., Dehmer, J. L., and Parr, A. C. (1998), "New ultraviolet radiometry beamline at the Synchrotron Ultraviolet Radiation Facility at NIST," Metrologia, 35, 301-306.

Shaw, P. S., Lykke, K. R., Gupta, R., O'Brian, T. R., Arp, U., White, H. H., Lucatorto, T. B., Dehmer, J. L., and Parr, A. C. (1999), "Ultraviolet radiometry with synchrotron radiation and cryogenic radiometry," Appl. Opt., 38, 18-28.

Tarrio, C., Grantham, S., Squires, M. B., Vest, R. E., and Lucatorto, T. B. (2003), "Towards high accuracy reflectometry for extreme-ultraviolet lithography," J. Res. Natl. Inst. Stand. Technol., 108, 267-273.

Tarrio, C., Grantham, S., Vest, R. E., and Liu, K. (2005), "A simple transfer-optics system for an extreme-ultraviolet synchrotron beamline," Rev. Sci. Instrum., 76, 046105.

USC, "SOHO SEM Instrument Home Page" (University of Southern California - Space Sciences Center, 2000), retrieved Dec 2004, http://www.usc.edu/dept/space_science/sem.htm.

Vest, R. E., and Canfield, L. R. (1997), "Photoemission from silicon photodiodes and induced changes in the detection efficiency in the far ultraviolet," AIP Conf. Proc., 417, 234-240.

Vest, R. E., Canfield, L. R., Furst, M. L., Graves, R. M., Hamilton, A., Hughey, L. R., Lucatorto, T. B., and Madden, R. P. (1999), "NIST programs for radiometry in the far ultraviolet spectral region" in Ultraviolet Atmospheric and Space Remote Sensing: Methods and Instrumentation II, Proc. SPIE 3818 (Eds, Carruthers, G. R. and Dymond, K. F.), pp. 15-26.

Vest, R. E., and Grantham, S. E. (2003), "Response of a silicon photodiode to pulsed radiation," Appl. Opt., 45, 5054-5063.

Woods, T. N., (2004) Personal communication.