Experimental setup

2.3.1 He(I) discharge lamp

He(I) radiation (21.22 eV) was generated using a commercial discharge lamp¹¹ (Omicron, HIS13), it can couple to a linear polarizer to generate a polarized photon source or a capillary to derive the unpolarized VUV light into the ionization chamber.

23

(A) He(I) with polarizer

Figure 2.5(a) shows the He(I) lamp with a continuously rotatable three-mirror linear polarizer. The polarizer consists of three gold coated mirrors with the special feature of additional focusing via a toroidal mirror,¹² the beam diameter is around 1.5 mm at the sample position, the polarization direction can be rotated 360 degree on VUV light path axis by adjusting the rotary drive (see Fig. 2.5), the maximum polarization efficiency is 88 % at the He(I) wavelength of 58.4 nm (21.22 eV). Due to the transmission of the polarizer is 4 %,

Figure 2.5(b) shows the He(I) lamp after removed the polarizer. In this case, 0.8-mm-inner diameter capillary was used to derive the VUV light beam into the ionization chamber and the photon source is unpolarized light. Accordingly to the manual, the beam divergence is estimated to be  0.8, and the beam diameter at the sample position can be estimated by the following equation

Beam dia. = 0.0272  L + (capillary inner dia.) (2-14) where L is the capillary–sample distance. Without the loss of photon flux caused by a reflective polarizer and an aperture in the photon beam, the photon flux is estimated to be 1.3

 10¹² photons/sec.

2.3.2 Vacuum chamber

The research works reported in this thesis were performed in two different vacuum chambers. In the beginning, the He(I)-PEI was tested with a vacuum chamber that was originally designed for laser spectroscopy of liquid droplets in our laboratory. However, after

24

some experiments, we confirmed that the electrostatic lens should be redesigned but that the size of this vacuum chamber geometrically restricts the modification. Therefore, we decided to use another vacuum chamber, with a larger inner space, designed for photoelectron imaging using a vacuum ultraviolet (VUV) free electron laser. In fact, the present work aimed at development of a PEI spectrometer that can be used with a VUV free electron laser, and this chamber was available for research and development of He(I) PEI for a plenty of time, since the beam time of the VUV laser was highly limited owing to time-sharing by the users. The use of this chamber turned out to be highly successful.

(A) The old vacuum chamber for laser spectroscopy of liquid droplet

As shown in Fig. 2.6, the apparatus consists of four vacuum chambers: a molecular beam source, a buffer, an ionization and a quadrupole chamber. A continuous supersonic jet was generated by expanding a sample gas through a 25-m-diameter orifice at 294 K in the source chamber that was evacuated by two turbo molecular pumps (2  2200 L/sec, BOC Edwards, F2203C). The supersonic jet was skimmed by a conical Al skimmer (2 mm ) in the source chamber, and collimated by the second conical Al skimmer (5 mm ) in the buffer chamber evacuated by a turbo molecular pump (350 L/sec, Leybold, TurboVAC361). The molecular beam thus generated was introduced into an ionization chamber evacuated by a 2650 L/sec turbo molecular pump (BOC Edwards, XA2703C), it travels parallel to the face of the imaging detector and intersected perpendicularly with He(I) radiation at 380 mm downstream from the nozzle. The differential pumping vacuum system provides the inside pressures of 3.1

 10^-5 Torr, 3.9  10^-6 Torr and 3.5  10^-8 Torr in the source, buffer and ionization chambers, respectively, during stagnation pressure of 0.9 MPa of Ar is operated.

Figure 2.7(a) shows a cross section of the ionization chamber. The He(I) discharge lamp with a polarizer was used. (see Sec. 2.3.1 A) The electrons generated by the photoionization of atoms or molecules in a beam were accelerated in static electric fields and projected onto a

25

2D position sensitive detector. The acceleration field was created using stacked circular ring electrodes and the field gradient was carefully adjusted to achieve a velocity mapping condition.³ This electrodes stack consist of eight stainless steel rings, which are spaced by 3 mm  ruby balls, and connected to each other by 1M resistors, as shown in Fig. 2.7(b). The whole regions of photoionization, acceleration, and flight for photoelectrons were shielded against an external magnetic field with a permalloy tube. After photoionization, the photoelectrons are accelerated perpendicular to both the propagation directions of a molecular beam and the VUV light, and detected by the position sensitive detector, consists of a dual microchannel plate (Hamamatsu F1942-04, 25 m  pore size, 25 m pitch length, and 77 mm  effective area), a phosphor screen (Hamamatsu P43), and a CCD camera (Hamamatsu i-CCD, 25 Hz, 768  572 pixels). Room light is blocked from the collection zone between phosphor screen and CCD camera.

(B) A new vacuum chamber

Figure 2.8 shows the schematic of the new vacuum chamber. The apparatus consists of two vacuum chambers, a molecular beam source and an ionization chamber. We used a continuous or pulsed molecular beam depending on the experiment. In the former case, a continuous supersonic jet was generated by expanding a sample gas from a 25-m-diameter orifice at 294 K. The jet was skimmed with a 2.0-mm-diameter skimmer and introduced into an ionization chamber, and a molecular beam is travelling parallel to the flight tube axis. The source and ionization chambers were pumped by turbomolecular pumps with pumping rates of 2000 L/s (N₂) and 820 L/s (N₂), respectively. When Ar was continuously expanded at a stagnation pressure of 1 atm, the ionization chamber pressure was ca. 7.3  10^-8 Torr. The polarizer of He(I) was removed (see Sec. 2.3.1 B), the end of the capillary was located ~125 mm from the ionization point. Consequently, the diameter of the He(I) radiation is estimated to be 4.2 mm at the ionization point by utilizing equation (2-14) and the photon flux is 9.4 

26

10¹⁰ photons/s/mm² at 21.22 eV (58.4 nm). Electrons generated by photoionization of a sample were accelerated along the molecular beam propagation axis and projected onto a 2D position-sensitive detector. The details of the electrodes are described later (see. chapter 3).

The entire regions of photoionization, acceleration, and flight of the electrons were shielded against external magnetic fields by a permalloy tube. The position-sensitive detector consists of a chevron-type (dual) microchannel plate assembly (Hamamatsu, F1942-04; pore diameter:

25 m; diameter: 77 mm) backed by a phosphor screen (P43) and a CCD camera (Andor, iXon^EM DCL897; 512  512 pixels).

In the time-gated experiments, 10–20% sample gases seeded in He were expanded from a pulsed valve (General Valve; orifice diameter: 100 m) with a stagnation pressure of about 0.2–0.55 atm. The front surface of the MCP assembly was maintained at the ground potential and the voltage applied to the rear side was switched between 900 and 1400 V using a high voltage pulser (DEI, GRX-3.0K-H; voltage amplitude: 3.0 kV) to time gate the MCP synchronously with a pulsed gas nozzle. The timing of the trigger pulses for the gas nozzle, MCP, and CCD camera were controlled with a digital delay generator (Stanford Research image must be found. This is first checked by eye or by overlaying a circle on the raw image to determine the center pixel. After an initial center is found, the pixels surrounding this center are also utilized as a center to reconstruct the full 3D distribution. Consequently, the corresponding radial distributions could be obtained from the angle integration of their 2D