Hatano Y., Katsumura Y., Mozumder A. (Eds.) Charged Particle and Photon Interactions with Matter - Recent Advances, Applications, and Interfaces

Подождите немного. Документ загружается.

50 Charged Particle and Photon Interactions with Matter

produced in their excited states and emit radiation, they are easily detected. But, when the fragments

are in the ground or metastable states, they cannot be easily detected. Two examples of the detection

neutral fragments are given as follows:

3.3.4.2.1

Neutral Radical Measurement for CH

All electronically excited states of CH

are repulsive states, leading to dissociation. This implies that

in any plasma containing CH

, chemically active neutral molecules (or radicals) are formed with

high efciency. For example, we expect to have CH, CH

, and CH

in large quantities. However, it

is very difcult to measure these neutral fragments. Tanaka and his group have developed a new

technique to measure an absolute cross section for the production of CH

through electron-impact

dissociation of CH

. This method is based on the combination of the crossed-beam method and the

threshold ionization technique (for details, see the paper Makochekanwa etal., 2006). Figure 3.23

(Makochekanwa etal., 2006) presents the cross section measured. In this gure, the threshold for

the neutral CH

formation is observed at 7.5 ± 0.3eV. Comparing with photoabsorption data, the

authors conclude that all the excitations of CH

below 12.5eV predominantly result in dissociation

via

the production of CH

3.3.4.2.2

Production of OH Radicals in H

In radiation interaction with matter, OH radicals are of particular importance. It is a fundamental

question, therefore, that how many OH molecules are produced when electrons collide with water

molecules. Harb etal. (2001) measured the cross section for the production of OH in its ground state

through electron-impact dissociation of H

O. They applied the laser-induced uorescence technique

to detect neutral OH. To obtain an absolute cross section, they compared the channel of produc-

tion H + OH to that of H

−

+ OH. The absolute cross section of the latter process is available in the

literature. The resulting cross section is shown in Figure 3.3. It is noted that, compared with other

processes,

the OH(X) production has a large cross section.

1.5

0.5

thr

: Present work

: CH

neutral diss., Kameta et al.

: CH

photoab., Kameta et al.

: CH

photoab., Au et al.

: Electron–photon

6 8 10

Impact energy(eV)

Cross sections (10

–16

)

Cross sections (10

–16

)

0.5

Figure 3.23 Absolute cross sections for the production of CH

radicals upon electron collisions with

. Also included are the photon impact results from the literature. For more details, see Makochekanwa

etal. (2006).

Electron Collisions with Molecules in the Gas Phase 51

3.3.4.3 negative-ion production

Electron interactions with biomolecules have been the subject of considerable interest, since it was

discovered that low-energy electrons can cause signicant DNA strand break (Boudaiffa etal.,

2000). Electrons with a specic energy readily attach to the molecular constituents of DNA to form

transient negative ions, which then dissociate to produce fragment negative ions and neutrals. This

bond-breaking process, called dissociative electron attachment, has been demonstrated to occur in

various

molecules, such as water, DNA bases, amino acids, and uorocarbons.

Dissociative

attachment is a kind of resonance process. It proceeds through a (doubly) excited

state

of the negative molecular ion in a way as follows:

e AB AB A+ B+ → →

− −

( )

The intermediate negative-ion state is unstable and called the “resonance state.”

Figure 3.24 shows one representative scheme of the DEA. In this case, the negative-ion state is

repulsive and crosses to the ground state of the neutral molecule at the internuclear distance R = R

When R is smaller than R

, the negative ion is unstable against the autodetachment of the electron

(i.e., AB

−

→ AB + e). Once R exceeds R

, dissociation to A + B

−

takes place automatically. When

theneutral molecule AB is initially in its vibrationally ground state, the attachment occurs only in

the Franck–Condon region. Depending on the steepness of the repulsive curve, dissociative attach-

ment has a nite value of the cross section only in a very narrow range of electron energy. When the

negative-ion state has an attractive potential, dissociative attachment can occur through the excita-

tion to the vibrational continuum of the negative ion.

Electron attachment to a molecule is studied by crossing an electron beam with a molecular

beam. The negative ions formed are recorded mass spectrometrically as a function of the electron

energy. Figure 3.25 (Itikawa and Mason, 2005) shows the cross sections of the dissociative attach-

ment to H

O to produce three different ions (H

−

, O

−

, and OH

−

). This is cited from the review article

by Itikawa and Mason (2005). After the publication of this article, a new measurement has been

reported by Rawat etal. (2007). Their results are a little, but not much, different from the values

shown

in Figure 3.25.

Franck–Condon

region

–

A+B

–

A+B + e

–

aff

diss

Capture cross section

Nuclear separation

Energy

Figure 3.24 Potential diagram of a diatomic molecule for the mechanism of dissociative attachment.

52 Charged Particle and Photon Interactions with Matter

One example of the DEA study of biomolecules is given here. Figure 3.26 shows the electron

energy dependence of the yield of a variety of negative-ion fragments, induced by a resonant attach-

ment of subionization electrons to thymine (Huels etal., 1998). To produce an effusive molecular

beam, high-purity (>99.5%) thymine is sublimated at 120°C–180°C, i.e., well below its decomposi-

tion temperature of about 320°C. The molecular beam is crossed at right angles with an electron

beam generated by a trochoidal electron monochromator, and the negative ions produced are mass-

analyzed by a quadrupole mass spectrometer. Table 3.1 (Huels etal., 1998) summarizes the peak

positions observed, and Figure 3.27 (Huels etal., 1998) shows the channels for the production of

each ion from gaseous thymine. This subject has been extended to the studies of fragmentation of

elementary compounds from condensed H

O, hydrated DNA, sugar analogues, oligonucleotides,

and

so on. For further information about these DEA fragmentations, see Sanche (2005).

3.3.5 colliSion proceSSeS involving vibrationally excited MoleculeS

In most of the experiments of electron–molecule collision, target molecules are in their (vibra-

tionally and electronically) ground states. In the nature, there are many molecules in their (par-

ticularly, vibrationally) excited states. Sometimes, the interaction between electrons and excited

molecules is signicantly different from that of the ground-state molecules, leading to a new subject

of study in atomic and molecular physics (Christophorou and Olthoff, 2001). One typical example

is the DEA. The DEA process is very sensitive to the internal degrees of freedom of the initial

molecule (Christophorou and Olthoff, 2001). The distribution of the rotational–vibrational states of

molecules depends on the temperature of the molecular gas. Hence, the cross section of the DEA

depends sensitively on the temperature of the target molecular gas (Christophorou et al., 1984;

Ruf etal., 2007). For other collision processes, little has been known about the target-temperature

dependence of the cross section. It is usually difcult to perform any experiment of electron colli-

sions with vibrationally excited molecules. This is mainly due to the difculty in producing excited

–17

–18

–

e+H

diss. attach.

–

–19

–20

–21

0 2 4 6

Electron energy (eV)

Cross section (cm

)

8 10 12 14

Figure 3.25 Dissociative attachment cross sections for H

O. Partial cross sections for the production of

−

, O

−

, and OH

−

are shown. (Reprinted from Itikawa, Y. and Mason, N.J., J. Phys. Chem. Ref. Data, 34, 1,

2005.

With permission.)

Electron Collisions with Molecules in the Gas Phase 53

species in a sufcient number density. Vibrationally excited molecules can be produced by heating,

laser

photoabsorption, or electron bombardment of the molecule.

The

rst quantitative experiment to study low-energy electron scattering from vibrationally

excited CO

was performed by Buckman etal. (1987). They measured the TCSs as a function of

temperature using a linear attenuation TOF spectrometer. They observed a substantial increase in

the TCS at electron energies below 2eV, which they attributed to the enhanced scattering by a bent

. No signicant change in the TCS was observed at higher energies. Ferch etal. (1989) repeated

the experiment of Buckman etal., also using a TOF spectrometer, and conrmed the earlier result.

They also found a signicant change in the resonance structure of the TCS at around 3–5eV, due to

vibrational

excitation of the target molecule.

DCS measurement for an electron collision with hot CO

was performed by Johnstone etal.

(1993), but only at one energy point (4eV). Johnstone etal. (1999) repeated the experiment at 3.8eV.

Recently, a similar but more extensive experiment was performed by Kato etal. (2008b). They cov-

ered the collision energies in the range of 1–9eV. This includes the resonance region around 3–5eV.

100

1.2

0.8

0.4

0.0

0.6

0.4

0.2

0.0

0 5 10

–

from

thymine

–

from

thymine

[T]

–

from thymine

(×10)

(a)

(b)

(c)

Incident electron energy (eV)

Mass-selected negative ion yield (10

counts)

Figure 3.26 Yields of representative negative ions produced in the electron irradiation of gas-phase

thymine, as a function of incident electron energy. Panel (a) shows the yield of the negative ion of the

parent molecule (thymine), panel (b) is the CN-yield, and panel (c) is the O-yield. (Reprinted from Huels,

M.A. etal., J. Chem. Phys., 108, 1309, 1998. With permission.)

54 Charged Particle and Photon Interactions with Matter

They obtained absolute cross sections for the inelastic (i.e., vibrational excitation) and the super-

elastic (i.e., vibrational de-excitation) transitions of CO

in its rst excited state of the bending mode

of vibration (as shown in Figure 3.28). The resonance structure of these cross sections has been

revealed for the rst time. This may be helpful in theoretically understanding the resonance in the

electron–molecule collision. Further extension of this kind of study is desirable also from the point

of view of application. It is well known that vibrationally excited molecules are abundant in the

upper atmosphere of the Earth and other planets and discharge plasmas.

table 3.1

peak

ositions

(in e

with u

ncertainty

of ±0.15

ev)

of the y

ield

of n

egative

Fragment

ons

from electron Collisions with t

hymine

and Cytosine

thymine (t)—C

–126 amu

−

oCn

−

oCnh

−

oCnh

−

0.18 0.20* 0.28* 1.8

3.4 2.8* 2.8* 3.2

4.6 4.8 4.6 4.5 4.0 4.6

5.2 5.8* 6.0* 5.2

6.4 6.6 6.6 6.6

7.7 7.8 8.0 8.7 8.0*

Cytosine

(C)—C

o–111 amu

−

oCn

−

or nh

−

or C

−

(0.1*)1.4 1.0 (0.1)1.4 1.5 1.2 2.3*

3.8 4.5

5.3 5.0 5.6 5.3* 5.1 5.2 5.2

6.9* 7.4 7.3 7.3*

7.7 7.8 8.2*

9.5 9.2

Source: Reprinted

from Huels, M.A. etal., J. Chem. Phys., 108, 1309, 1998.

With

permission.

Note:

The

asterisks indicate a faint peak or a shoulder.

7 reaction pathways with

7 anion and associated

neutral fragments

OCN

–

+ {C

NO + H}

–

+ {C

NO + O + H}

OCNH

–

+ C

OCNH

–

+ C

–

+ C

–

+ C

–

+ {C

+ H}

+ {C

+ H + H}

–

Figure 3.27 Pathways of producing negative fragment ions from electron collisions with thymine.

(Reprinted

from Huels, M.A. etal., J. Chem. Phys., 108, 1309, 1998. With permission.)

Electron Collisions with Molecules in the Gas Phase 55

3.3.6 coMplete colliSion dynaMicS in (e, 2e) experiMentS for h

As described in Section 3.3.4.1, partial and total ionization cross sections are of great importance

for practical applications, but these provide only limited insight into the ionization process itself.

Conversely, a great deal of information may be obtained from the study of DCSs (Märk, 1984), that is,

SDCS :

,d T W

( )

DDCS :

( , , )d T W

dW d

σ θ

Ω

Triple DCS :

( , , , , )d T W

dW d d

1 2 2

1 2

σ θ θ ϕ

Ω Ω

where

is the energy

and φ are the polar angles of detection of the electrons after collision (see Figure 3.29; Märk, 1984)

In addition to this electron–electron spectroscopy, an ion kinetic energy spectroscopy and the mea-

surement of angular distributions of fragment ions are necessary for a full picture of the ionization

event. It is outside the scope of this chapter to give a detailed description on the subject, but a brief

comment

about the recent advances is given here.

0.25

0.20

0.15

Angle=30°

(a) (b)

0.10

0.05

0.00

0.20

0.15

0.10

0.05

0.00

0.25

0.20

0.15

0.10

0.05

0.00

0.20

0.15

0.10

0.05

0.00

1 2 3 4 5 6

Impact energy(eV)Impact energy(eV)

DCS (10

–16

/sr)

DCS (10

–16

/sr)

7 8 9 1 2 3 4 5 6 7 8 9

Present work

Johnstone et al.

Antoni et al.

Present work

Johnstone et al.

(000)

(010)*

(010)

(010)*

(000)

(020)(100)

(020)*

(100)* (010)*

Present work

Johnstone et al.

Present work

(010)*

(000)

(020)*

(100)* (010)*

Angle=60°

Present work

Johnstone et al.

Antoni et al.

Present work

Johnstone et al.

(000) (010)

(010)*

(020)(100)

Figure 3.28 Vibrational excitation functions for the inelastic (000)→ (010) and (010) → (020) (100) and

the superelastic (010) → (000) and (020) (100) → (010) transitions of CO

at impact energies of 1–9 eV and at

scattering angles of (a) 30° and (b) 60°. See Kato etal. (2008b) for further details. (Reprinted from Kato, H.

etal.,

Chem. Phys. Lett., 465, 31, 2008b. With permission.)

56 Charged Particle and Photon Interactions with Matter

Electron momentum spectroscopy (EMS), also known as (e, 2e) spectroscopy, has made a sub-

stantial contribution to the study of the electronic structure of matter. The basic principle of EMS is

a complete measurement of the momenta of the two electrons emerging after collision. It is a kine-

matically complete study of electron-impact ionization events. Figure 3.30 shows typical examples

of the results of the (e, 2e) experiment for gaseous water (Haed etal., 2007). This experiment was

performed in a symmetric noncoplanar geometry. A high-energy incident electron (E

= 1200 eV)

knocks out one electron from H

O, and the two outgoing electrons are detected in coincidence at the

Figure 3.29 Schematic view of the kinematics of an electron-impact ionization: incident electron e

with

kinetic energy T

and momentum k

; and scattered and ejected electrons e

and e

with kinetic energies W

and W

and momenta k

and k

, respectively. (Reprinted from Märk, T.D., Ionization of molecules by elec-

tron impact, in Christophorou, L.G. (ed.), Electron-Molecule Interactions and Their Applications, Vol. 1,

Academic

Press, New York, 1984, 251–334. With permission.)

8×10

–4

6×10

–4

4×10

–4

2×10

–4

1×10

–3

8×10

–4

6×10

–4

4×10

–4

2×10

–4

4×10

–3

3×10

–3

2×10

–3

1×10

–3

8×10

–4

6×10

–4

4×10

–4

2×10

–4

0.0 0.5 1.0 1.5 2.0 2.5

0.0 0.5 1.0 1.5 2.0

Momentum(a.u.)

TDCS (a.u.) TDCS (a.u.)

Momentum(a.u.)

2.5 0.0 0.5 1.0 1.5 2.0 2.5

0.0 0.5 1.0 1.5 2.0 2.5

Figure 3.30 EMS momentum proles for the four valence orbitals of gas-phase H

O. Full dots with error

bars are the experimental data taken from Bawagan etal. (1987). Lines are the results of calculations using

different levels of wavefunctions. (Reprinted from Haed, H. etal., Chem. Phys. Lett., 439, 55, 2007. With

permission.)

Electron Collisions with Molecules in the Gas Phase 57

same high energies (E

= E

≃ 600eV) and the same polar angles (θ

= θ

= 45°). The experimental

result (Bawagan etal., 1987) is compared with several quantum chemical calculations. All the theo-

retical proles in Figure 3.30 have been folded with the instrumental momentum resolution. The

results of these studies have been applied practically to the charged-particle track structure analysis

both vapor and liquid waters.

The

(e, 2e) experiments, however, only give the result averaged out over molecular directions.

Of late, Takahashi and his colleagues (Takahashi etal., 2005) have started a new experiment that

provides the information of a molecule xed in space. They measure the momentum of a fragment

ion in a dissociative ionization, in coincidence with the two outgoing electrons (i.e., a triple coinci-

dence experiment). In Figure 3.31, the scheme of the measurement is given. For further details, see

Takahashi etal. (2005).

3.4 appliCations oF Cross-seCtion data and databases

3.4.1 energy depoSition Model baSed on colliSion croSS SectionS

The aim of the study of electron–molecule collisions is not only to learn the physics of collision

dynamics but also to provide a set of recommended data of the cross section for various practical

applications. As an example of applications of the cross-section data, a model of energy deposition

of radiation in water vapor is shown here. As is mentioned earlier, it is understood now that even

sub-ionizing electrons can produce damages (in terms of strand breaks and molecular dissociation).

We need to consider collision processes over a very wide range (from meV to MeV) of electron

energies.

Garcia and his colleagues have developed a new simulation program based on the Monte

Carlo method (Muñoz etal., 2008). For this program, they prepared an extensive set of cross-

section data for elastic-scattering and inelastic (ionization and excitation) processes. They also

–

, p

)

–

, p

)

–

, p

)

φ=0°

DMD

Δφ

CEMs

EAs

H H

Figure 3.31 Schematic diagram of an (e, 2e + M) experiment for a symmetric noncoplanar geometry.

This shows seven channel electron multipliers (CEMs), and a pair of entrance apertures (EAs) of a spherical

analyzer with a pair of position-sensitive detectors. (Reprinted from Takahashi, M. etal., Phys. Rev. Lett., 94,

213202-1,

2005. With permission.)

58 Charged Particle and Photon Interactions with Matter

used the TCS to estimate the mean free path of incoming electrons. When no comprehensive data

were available experimentally, they resorted to theoretical calculations. For instance, the angular

dependence (i.e., DCS) of the elastic cross section was derived from a model potential calculation.

Figure 3.32 shows one sample result of their simulation of 5 keV electron tracks in 1 atm water

vapor. This shows one aspect of the track structure: local distribution of energy deposition events.

They can identify each event with a specic collision process. They plan to extend this model to

water in the liquid phase.

3.4.2 electron tranSport in high-preSSure xe doped with ch

High-pressure Xe has been studied extensively in relation to the development of a new-generation

gamma-ray camera for applications in planetary science (Pushkin etal., 2006), as well as of com-

pact and faster versions of radionuclide imaging in nuclear medicine (Barr etal., 2002). As a more

familiar example, the Xe gas has been commonly used in the plasma display panel (PDP) (Sobel,

1998) for producing, efciently and steadily, near-UV light (at a wavelength of 147nm or photon

energy of 8.43eV). The light is emitted as a result of a transition from the resonant state to the

ground state (5P

[

3/2

] 6s→5P

[

]). For lowering the discharge voltage, a mixture of Xe and a

few

percent of He or Ne is used to produce metastable states of He

and Ne

developing the next-generation gamma-ray imager (Pushkin etal., 2006), it was found that the

electron diffusion and scintillation properties in a high-pressure Xe gas (Hunter and Christophorou,

1984) are greatly inuenced by adding molecules such as CH

and H

. Scintillation in xenon, initi-

ated by a Compton electron, is used to generate a start signal, which enables the point of interaction

of the γ-ray quantum with the detector material to be reconstructed by the drift time of the ejected

secondary electrons. Xenon is characterized by a low drift velocity of “hot” electrons, which limits

the diffusion of the electron cloud drifting toward the electrode (the anode). An addition of methane

(∼0.2%) to xenon signicantly increases the drift velocities of electrons, as shown in Figure 3.33.

1.5

0.5

–0.5

y (mm)

–1

<50 eV

50–100 eV

>100 eV

–1.5

0.50 1 1.5

x (mm)

2 2.5 3

Figure 3.32 Simulation of 5keV electron tracks in 1atm of water vapor. Three different amounts of energy

deposition are separately indicated. (Reprinted from Muñoz, A. etal., J. Phys.: Conf. Ser., 133, 012002-1,

2008.

With permission.)

Electron Collisions with Molecules in the Gas Phase 59

Thus, a doping of methane helps in improving the time and coordinate characteristics of the Xe time

projection

chamber counter.

3.4.3 databaSeS

Electron–molecule collisions have been extensively studied for a long time. Cross sections have

been reported in a wide range of publications. Sometimes it takes time to survey the literature to

nd suitable data. It is therefore useful to compile cross-section data and publish them. A list of data

compilations of electron–molecule collision cross sections is given in the book by Itikawa (2007).

This

book also summarizes the methods to nd the cross-section data.

One

of the most comprehensive compilations of cross-section data for electron–molecule colli-

sions has been published as a volume of the Landolt–Börnstein series of data books (Itikawa, 2003).

This includes the cross sections for ionization, electron attachment, total scattering, elastic scatter-

ing, momentum transfer, and excitations of rotational, vibrational, and electronic states of more than

70 molecular species. For individual species of a molecule, Itikawa and his colleagues published a

series of data compilations in the Journal of Physical and Chemical Reference Data (Itikawa, 2002,

2006,

2009; Itikawa and Mason, 2005; Yoon etal., 2008).

Almost

all the compilations of cross-section data published so far are the collection of an integral

cross section (i.e., the cross section integrated over scattering angles). As is emphasized earlier, the

Monte Carlo simulation of radiation tracks needs the information of the scattering angle distribution

of electrons (i.e., the DCS). Normally, the numerical data on DCS are too detailed to be presented

in a compact form. Tanaka and his group recently published a collection of DCS for elastic scatter-

ing of electrons from polyatomic molecules, which they experimentally obtained (Hoshino etal.,

2008). This paper provides at least a part of the information needed in the Monte Carlo simulation.

Those who use the databases should notice one thing: any recommended values of cross-section

data can be changed due to the development of experimental techniques and theoretical methods.

One

should be careful to use the most recent version of the databases.

Drift velocity of electrons, W, 10

cm/s

Xe+ CH

(0.2%)

Xe+ CH

(0.4%)

Xe+ CH

(0.6%)

Xe+ CH

(1.0%)

Pure Xe

0 0.1 0.2 0.3

E/N, 10

–17

V cm

0.4

Figure 3.33 Drift velocities of electrons in pure gaseous xenon and xenon–methane mixtures with differ-

ent methane concentrations (0.2%–1%), as a function of reduced electric eld E/N. (Reprinted from Pushkin,

K.N.

etal., Instrum. Exp. Tech., 49, 489, 2006. With permission.)