Amendment 2.docx

Amendment 2: Answers to Questions

Questions:

Firstly, we have difficulties to understand the bandwidth specifications in 1.2.

Is one of these two definitions a typo?

If not, can you please explain how “bandwidth 10% FWHM” is meant.

S1.2) Photon flux : i) The VUV+EUV source shall have the following minimum photon flux per wavelength and specified spectral bandwidth (full width at half maximum = FWHM). Metric applies to the output at the source, taking into consideration known and anticipated losses inherent in wavelength selection: (1) Energy 92 eV, (13 nm wavelength), spectral bandwidth 10% FWHM: photon flux >= 1E10 photons per second

(2) Energy 40 eV, (31 nm wavelength), spectral bandwidth 5 eV FWHM: photon flux >= 1E10 photons per second.

If the flux is specified within 5eV around 40eV this will contain multiple harmonic lines. Is that wanted?

Secondly, if this is the way it is meant, why would the flux be specified within 5eV around 40eV when later on it is cropped to 0.4eV around 40eV anyways according to the spec in 2.6.

Can you clarify?

S2.6) The VUV+EUV source shall include a monochromator that will enable the selection of wavelengths at the target. The monochromator shall be a component of the VUV+EUV Source. i) The monochromator component of the VUV+EUV Source shall have a minimum spectral bandwidth no greater than 1 % the wavelength (full width at half maximum (FWHM)) across the 10 nm to 150 nm wavelength range.

ii) The specification above may require one or multiple gratings, to be included.

Answers to Questions:

Addressing these two questions with a combined response. For a polychromatic source filtered by bandwidth, measurements of the photon flux at a specified wavelength will depend on the energy-dependent output of the source and the filtering used. “Full width at half maximum” (FWHM) is widely used for visible optics with transmissive bandpass filters, which typically yield a Gaussian distribution of wavelength, and such distributions must specify at what height the width is to be measured. A major difference between the extreme ultraviolet (EUV) and the visible is that filtering of wavelengths (energies) is typically performed using a grating and a set of slits or equivalent to allow a range of wavelengths (energies) to pass through. Momentarily assuming any sort of EUV broadband source, the use of slits would favor more of a rectangular distribution than a Gaussian distribution, but FWHM for a rectangular distribution is simply its full width.

From our market research, we are aware of several technologies to produce EUV light. Specifications S.1.2.i.(1) and S.1.2.(2) are intended to make clear that a photon flux of at least 10^10 photons/sec is required for our application for at least two different crucial energies. Practically, each EUV light source technology is non-uniform in amplitude with respect to energy and it is to be specified how bandwidth is to be defined. From S.1.2, this source shall yield 10^10 photons both at 92 eV and 40 eV, but with different permitted bandwidths, up to 10 % FWHM for 92 eV and 5 eV for 40 eV. This can be interpreted also as 9.2 eV FWHM for 92 eV and 5 eV for 40 eV. Thus, if the underlying source is uniform in intensity across all wavelengths and slits are used in the EUV, then the resulting rectangular distribution would permit all source energies between 87.4 eV to 96.6 eV and between 37.5 eV to 42.5 eV to be combined to demonstrate the required combined flux of 10^10 photons/sec. Note that if the underlying source is not uniform and there is a roll-off of intensities greater than half the maximum amplitude, then this range would narrow.

It is difficult to assess photon flux across all potential technologies that exist in the EUV thus S.1.2 was written to ensure that this source would be bright for at least two specific energies that can be judged across technologies. With regards to high-harmonic generation (HHG), this is one of the several EUV technologies. One part of the question reads “If the flux is specified within 5eV around 40eV this will contain multiple harmonic lines. Is that wanted?” With HHG it is understood that the intensity distribution would not be a continuous output of energies at 40 eV and 92 eV but rather a comb of harmonics. Specification S.1.2.(2) is written to allow consideration of the sum of all harmonics between 37.5 eV to 42.5 eV to be included in the consideration of photon flux which must be 10^10 photons/sec if the amplitude of each of these is above one-half of the maximum intensity. This is also true for S.1.2.(1). It is accepted that an HHG will produce a comb of energies convoluted together in this energy range, but that in general this can be treated like a rectangular distribution for determining the bandwidth FWHM. Roll-off in intensities of harmonics might narrow this range.

As to the second question: “Secondly, if this is the way it is meant, why would the flux be specified within 5eV around 40eV when later on it is cropped to 0.4eV around 40eV anyways according to the spec in 2.6. Can you clarify?” S1.2 was written to provide a specification that large photon flux is required at not just one specific range of energies but at two separate regions of energy space. To further clarify the specifications needed for the source, we specified S2.6 – that “minimum spectral bandwidth no greater than 1 % the wavelength (full width at half maximum (FWHM)) across the 10 nm to 150 nm wavelength range.” This is the minimum bandwidth that the source shall produce during normal operation. It logically follows that that the source near 40 eV with this new minimum will likely not yield 10^10 photons at this minimum bandwidth. While harmonics near 40 eV may be well-separated in energy, from our market research it appears that these harmonics cluster at higher energies. Note the use of “near 40 eV” as it is understood that drive laser energy will determines the precise harmonic(s) available near 40 eV.