More than a dozen ions hitherto identified in the interstellar medium.

Interstellar chemistry once thought to be dominated by ion chemistry.

Ions found in interstellar clouds, shock waves, ionospheres, etc.

The “Horsehead nebula” (Ori) The “Cat’s eye” (Draco) Aurora over Alaska

Interstellar ion chemistry

Ion - neutral reactions ( H2+ + H2 H3

+ + H )

Ion - electron reactions ( H3+ + e- 3 H )

Ion - ion reactions (H- + H+ 2 H ) (quite unexplored)

Important ion reactionsin the ISM

Cosmic RayIonization

Ion-MoleculeReactions

Recombination

CH3+ + H2O CH3OH2

+

CH3OH2+ + e- CH3OH + H

For example: Presumed synthesis of methanol

1. Ion-molecule reaction

2. Dissociative recombination

Feasible pathway of moleculesynthesis in space

h

1rad

kc

1Dis

Redissociation in competition with radiating off of energy

CH3+ + H2O CH3OH2

+ + h

Scheme of a radiative association

Radiative association

AB + h Radiative

recombination(too slow)

AB+(v=m) + e-

Elastic/inelastic/superelastic

(n=m/n<m/n>m)scattering

A+ + B- Resonant ion

pair formation

(high energies) AB+

(v=n) + e-

A + B Dissociative

recombination

Important electron-ion reactions

Negative charge in the interstellar medium (ISM) thought to present mostly in the form of electrons.

DR often the final step of synthesis of neutral molecules in the ISM. (HX+ + e- X + H )

Dissociative recombination(DR) often the only way to destroy cations.

Ample data on of DR rates, little on branching ratios.

Branching ratios often hard to explain by ”common sense”.

DR can lead to excited states that emit characteristic lines.

Dissociative recombination (DR) in interstellar chemistry

Dissociative recombination (DR) in space

General rule: Conditions must match interstellar ones:

Ions have to be rotationally and vibrationally cool. Three-body processes must be excluded.

Low relative translational energies of reactants.

Additionally: Clear identification of the ion (isomeres) and products.

Up to the 90’s measurements restricted to afterglow experiments.

Problems to quantifyDR reactions

Bates’s theory 1986: Dissociative recombinatons favour the pathway(s) which involve(s) least orbital rearrangement, e. g.:

N2H+ + e- N2 + H N2OH+ + e- N2O + H

Difficult to obtain reliable potential surfaces due to involvement of highly excited states

very few high-level ab initio studies on DR reactions available

Theoretical prediction of the pathways of DR reactions

4 steps:

1. Production of He+ by discharge in He: He + e- He+ + 2 e- 2. Reaction of He+ with H2: He+ + H2 H2

+ + He H2

+ + H2 H3+ + H

3. Reaction of H3+ with other substances, e.g. CO:

H3+ + CO HCO+ + H2

4. Recombination of the ion: HCO+ + e- H + CO

Flowing afterglow

Glosik et al. 2006

+ Low operational costs.

+ Thermal equilibrium of reactants.

+ Detection of products by mass spectromtry.

+ Detection of electron degradation by Langmuir probe.

- Impure reactants - except very simple systems like H3

+.

- Mearurements only at high (room) temperatures.

Advantages and disadvantages of flowing afterglow

Schematic view of CRYRING

Steps during the experiment

1.. Formation of the ions in the source

2. Mass selection by bending magnet

3. Injection via RFQ and acceleration

4. Merging with electron beam

5. Detection of the neutral products

1

23

45

Storage ring (CRYRING)

Electron cooler

Bendingmagnets

Bendingmagnets

Cooledcathode

Anode

IonBeam Neutral

fragments

ProbabilityT(1-T)

with grid

without grid

Signal with grid(mass spectrum dependent

on branching ratio and T)

Signal without grid(all events lead to full

mass signal)

Surface barrierdetector

e-

GridT=0.3e-

Branching ratio

Particle loss

GRID technique

Low (interstellar) relative kinetic energies of the reactants.

Mass selection of the ion produced.

All products can be identified.

Low background.

Only radiative cooling possible.

No straightforward identification of product internal states.

High set-up and operation costs.

Advantages and disadvantagesof storage rings

One of the most prominent ions in dark interstellar clouds.

N2 lost through protonation might be fully recovered by DR of N2H+:

N2 + H3 N2H+ + H2 N2H+ + e- N2 + H

Most of interstellar nitrogen thought to be stored as N2.

Tracer for the unobservable N2.

Present in Titan’s ionosphere.Saturn’s satellite Titan

N2H+ + e-

The ”Red Rectangle”

HCO+ formed easily in the interstellar medium from CO through protonation (e. g. by H3

+).

One of the most important carbon- containing interstellar ions.

Cameron bands in Red Rectangle maybe due to excited CO from DR of HCO+.

Cameron bands in the Red Rectangle

HCO+ + e-

HCS+ is the most important sulfur-containing interstellar molecular ion.

In dark clouds, a high HCS+/CS ratio is found.

CS presumably formed by DR of HCS+.

Very low rate of DR used in astrophysical models.

How does the rate and branching ratio of the DR affect the HCS+/CS ratio ?

HCS+ + e-

N2H+ + e- N2 + H H = - 8.47 eV

NH + N H = - 2.40 eV

N2H+ + e- reaction channels

0 1 2 3

0

200

400

600

800

1000

1200

2N+H2N

N+HN

H

Sig

nal

Inte

nsi

ty /

cou

nts

Fragment kinetic energy / MeV

N2H+ fragment energy spectrum

HCO+ + e- H + CO (X 1S+) H = - 7.45 eV

H + CO (a 3P) H = - 1.43 eV

H + CO (a 3S+) H = - 0.75 eV

HC + O H = + 0.17 eV

OH + C H = - 0.75 eV

HCO+ + e- reaction channels

1.0 1.5 2.0 2.5 3.0 3.5

0

2000

4000

6000

8000

10000

12000

14000

16000

18000

Sig

na

l in

ten

sity

/ co

un

ts

Fragment kinetic energy / MeV

HCO+ fragment energy spectrum

C

C+H

O

O+H

C+O

C+O+H

N2H+ + e- / HCO+ + e-

GridT=0.3e-

N2H+ + e- N2 + H H = - 8.47 eV

NH + N H = - 2.40 eV

N2H+ + e- reaction channels2 2

2N H

2N2

N H

N

H

IT T

IT(1 T) 0BR(N H)

I0 T(1 T)BR(NH N)

I0 T(1 T)

IT(1 T) 0

Evaluation matrix

N2H+ + e- / HCO+ + e-

2 22N H

2N2

N H

N

H

IT T

IT(1 T) 0BR(N H)

I0 T(1 T)BR(NH N)

I0 T(1 T)

IT(1 T) 0

    

Branching ratios  

Branching ratioReaction channel

0.36N2 + H

0.64N + NH

Evaluation matrix

Evaluation of the branching ratios

Cross section of N2H+ + e-

k / cm3 mol-1s-1

FALP*Our data

1.7 10-7 (T/300)-0.90 1.0 0.110-7 (T/300)-0.510.02

Reaction rates of N2H+ + e-

•Taken from: Smith, D., & Adams, N. G. 1984, ApJ, 284, L13

Dependence of on relative kineticenergy

k(T)

3/ 21/ 2

kinkin kin kin

B B0

1 2 Ek(T) (E ) E exp dE

k T k T

N2H+ + e-

    

N2H+ + e- HCO+ + e- HCS+ + e-

Reaction channel

Branching ratio

Reaction channel

Branching ratio

Reaction channel

Branching ratio

N2 + H 0.36 ± 0.05 CO + H 0.92 ± 0.04 CS + H 0.20 ± 0.03

CH + O 0.01 ± 0.02 NH + N 0.64 ± 0.05

C + OH 0.07 ± 0.02

CH + S C + SH

0.80 ± 0.03

Branching ratios

Reaction rates

}

This work Previous FALP Reaction

Rate / cm3 s-1 Rate / cm3 s-1

N2H+ + e- 1.0 0.3 10-7 (T/300)-0.51 0.02 1.7 10-7 (T/300)-0.90 1

HCO+ + e- 3.0 0.9 10-7 (T/300)-0.74 0.02 2.4 10-7 (T/300)-0.69 2

HCS+ + e- 9.7 1.9 10-7 (T/300)-0.57 0.02 -

1 D. Smith, , & N. G Adams, Astrophys. J., 284, L13 (1984) 2 J. B.A. Mitchell, Phys. Rep., 186, 5 (1990)

N2H+ + e- / HCO+ + e-

R / au

HCO+ is depleted in the centre of the core, N2H+ is constant, NH3

slightly enhanced.

Explanation: CO frozen out, N2 isn´t.

Aikawa et al. 2005

N2H+ in prestellar cores

BUT

Temperature desorption behaviour of N2 and CO differs only slightly. (Schlemmer and co-workers 2006)

No explanation for enhancement of ammonia near the core centre.

Explanation

Two destruction mechanisms for N2H+, only DR for HCO+:

N2H+ + CO HCO+ + N2

N2H+ + e- Products At low temperatures DR becomes the only degradation process (CO frozen out, but N2 also)

Formation of NH leads to enhancement of NH3.

Taken from Aikawa et al. 2005

Beamsplitter

MCP

Phosphorus screen

PMP

CCDcamera

Trigger

e-

v

Can we gather information about the product kinetic energy ?

Imaging analysis

HCO+ + e- H + CO (X 1S+) H = - 7.45 eV

H + CO (a 3P) H = - 1.43 eV

H + CO (a 3S+) H = - 0.75 eV

HC + O H = + 0.17 eV

OH + C H = - 0.75 eV

HCO+ + e- reaction channelsReaction channels leading to differentelectronic energy levels of CO

HCO+ + e-

0 10 20 30 40

0

1000

2000

3000

4000

5000

6000

7000

CO - D distance / mm

X1+ + a3

X1+ + a3 + a'3+

exclusively X1+

Cou

nts

Exp. data

X1+ 30%

a3 44%

a'3+ 26%

X1++a3+a'3+

Fit of the different electronic state contributions

Imaging of DCO+

In the DR of N2H+, the break-up of the N-N bond dominates.

In the DR of HCO+, the CO + H channel is preeminent.

Recombination of HCO+ partly leads to CO in the 3Pu state, which can explain the Cameron bands in the Red Rectangle.

In the DR of HCS+, the break-up of the C-S bond is favoured.

Reaction rate in the N2H+ and HCO+ DR reactions in agreement with previous FALP measurements.

Conclusions N2H+/ HCO+/ HCS+

*Ohishi, M. , Irvine, W. M. Kaifu, N., Astronomy of Cosmic Phenomena, 171

Abundances / N(species/N(H2)

Early time (3.16 105 yr)

Steady state Observed values

(TMC-1) Species

RATE99 This work RATE 99 This work Ohishi et al.*

N2H+ 6.24(-12) 1.23(-12) 1.14(-10) 3.39(-10) 1.77(-10)

NH3 3.59(-9) 2.39(-10) 3.61(-8) 6.73(-8) 2.00(-8)

HCN 2.99(-8) 4.76(-10) 3.42(-9) 8.43(-9) 2.00(-8)

HNC 5.39(-8) 6.85(-10) 2.77(-9) 5.32(-9) 2.00(-8)

OH 1.76(-8) 1.35(-8) 4.07(-8) 1.23(-7) 1.00(-7)

Some notorious troublemakers:

H2O 3.27(-6) 3.60(-9) 3.33(-6) 3.18(-7) 6.00(-10)- 2.00(-8)

HCOOH 1.67(-8) 4.50(-11) 4.00(-8) 6.30(-9) <2.00(-10)

New branching ratios in a model of TMC-1

Abundances of N-containing compounds predicted better assuming an older age of TMC-1.

Some improvements for molecule densities that proved difficult to model (H2O, HCOOH).

No big influence on models of circumstellar envelopes, planetary nebulae and diffuse clouds.

Conclusions from model calculations

Influence on interstellar sulfur chemistry.

SO2 is found in atmospheres of planets (Venus) and satellites (Io).

Important role of SO2+ in the ionosphere of Io.

Three-body break-up energetically allowed.

Iupiter´s moon Io

SO2+ + e-

SO2+ + e- reaction channels

SO2+ + e- SO + O H = 6.32 eV

S + O2 H = 6.40 eV

S + 2O H = 1.22 eV

Branching ratios of SO2+ + e-

Reaction rate

k(T) = 4.6 0.110-7 (T/300)-0.520.02 cm3 mol-1s-1

Reaction product

Branching ratio

SO + O 0.61 S + O2 0.00

S + O + O 0.39

SO2+ + e-

Decay of SO2+ in Io’s ionosphere during eclipse probably

caused by DR.

Strong observed UV lines of O(I) and S(I) could be due to increased S- and O-atom production by three-body break- up in DR.

Possible role in the ionosphere of Venus ?

Consequences

Responsible for maser emission in star-forming regions.

Evolution indicator in star-forming regions

Used for determination of kinetic temperature and H2

density simultaneously.

From CH3OH2+/CH3OH ratio

electron temperature in cometary coma derived.

The Bear Claw Nebula, where a strong methanol maser was detected

Methanol in space

CH3+ + H2O CH3OH2

+

CH3OH2+ + e- CH3OH + H

With a high rate of DR, the radiative association rate should be about 1.2 10-10 cm3s-1 at 50 K.

(Herbst et al. 1985)

Production of methanol in the ISM

Ion trap experiments yielded a an upper limit of 2 10-12 cm3s-1 at dark cloud temperatures (Luca et al.

2002).

a factor of 60 too low !

But...

However...

CH3+ not detected so far, densities only

estimates from models.

Uncertainties in water densities.

If the DR of CH3OH2+ leads to methanol with a

branching ratio of close to 100 %.......

0.0 0.5 1.0 1.5 2.0 2.5 3.0

0.0

2.0x103

4.0x103

6.0x103

8.0x103

1.0x104

1.2x104

1.4x104

D

2D

3D C+D

C+2D O

O+4D

C+3DO+D

C+4DO+2D

C+5D,O+3D

C+O+2D

C+O+3D

C+O+4D

C+O+5D

In

tens

ity

/ cou

nts

Fragment kinetic energy / MeV

Fragment energy spectrum of CD3OD2

+

CD3OD2+ + e- CD3OD + D

CD3 + OD + D

CD2 + D2O + D

CD + D2O + D2

CD3O + 2D

CD3O + D2

CD2O + D2 + D

CD2O +3D

CD4 + O + D

CD3OD2+ + e- CD4 + OD

CD2 + OD + D2

CD3 + D2 + O

CD3 + D2O

CDO + 2D2

CDO + D2 + 2D

CO + 2D2 + D

CO + D2 + 3D

Some of the channels deliver products with the same mass indistinguishable.

Fragment energy spectrum of CD3OD2

+

Reaction pathway

Branching ratio

CD3OD + D 0.06

CD3 + D2O (CD4 + OD)

0.11

CD3O + D2 0.05

CD3 + OD + D 0.59

CD2 + D2O + D (CD4 + O + D)

0.16

CD + D2O + D2 0.01

CD3O + 2D 0.00

CD2O + D2 + D 0.02

CD2O + 3D 0.00

Reaction pathway

Branching ratio

CH3OH + H 0.03

CH3 + H2O 0.09

CH4 + OH 0.00

CH3O + H2 0.07

CH3 + OH + H 0.51

CH2 + H2O + H 0.21

CH4 + O + H 0.00

CH + H2O + H2 0.00

CH3O + 2H 0.00

CH2O + H2 + H 0.09

CH2O + 3H 0.00

Branching ratios CD3OD2+/CH3OH2

+

Processes

Sum of branching

ratios (CD3OD2

+)

Sum of branching

ratios (CH3OH2

+)

2- body 0.22 0.19

3- body 0.78 0.81

4- body 0.00 0.00

2-,3- and 4-body processes

Cross-section vs. collision energy

10-3 10-2 10-1

10-15

1x10-14

1x10-13

1x10-12

1x10-11

Cro

ss s

ect

ion

/ cm

2

Relative kinetic energy (eV)

= 9.55 10-16 E(eV)-1.2cm2

Thermal reaction rate (CD3OD2

+):

k = 9.11 10-7 (T/300)-

0.63

cm3s-1

For the undeuterated isotopomer (CH3OH2

+):

k = 8.91 10-7 (T/300)-

0.59

cm3s-1

CD3OD2+

2-,3- and 4-body processes

CH3OH + H branching ratio = 1

CH3OH + H branching ratio = 0.06

Including new rates for the radiative association of CH3+ and H2O,

(Luca et al. 2002) the peak methanol relative abundance sinks to 7 10-13.

UMIST (Rate99) model predictions for methanol density in TMC-1

Observed methanol density (TMC-1)

Model Calculations

UMIST (Rate04) model predictions for methanol density in TMC-1

CH3OH + H branching ratio = 1

CH3OH + H branching ratio = 0.06

+ new rate forCH3

+ + H2O

Observed methanol density (TMC-1)

Main gas phase route to CH3OH is now CH3CHO + H3

+ CH3OH + CH3+

k = 1.4 10-9cm3s-1 at 300K

New UMIST model

Three-body break-ups dominate.

Production of CH3OH only 3 % (CD3OD only 6 %).

No big isotope effects

Gas-phase mechanism for interstellar methanol very unlikely.

In line with the following facts:

Formation of methanol on CO ice surfaces possible at 10 K. (Watanabe et al. 2004) Models including grain surface desorption reproduce methanol

densities (Herbst 2006)

Conclusions

Can we close the books ?

Anticorrelation of CO and CH3OH in dense clouds. (Buckle, 2006)

No experimental evidence for surface desorption of freshly formed methanol

DCO+ CD2OD+ CD2OD2+ CD3OD2

+

Break-up of C-O bond 0.08 0.08 0.43 0.81

C-O bond intact 0.92 0.92 0.57 0.19

Retention versus break-up of CO-bond

Increasing hydrogen saturation favours C-O bond rupture A rule for DR of hydrogen-containing ions ?

DR of other CHxO+ systems

Similar mechanism to methanol postulated for dimethyl ether.

Similar problems ?

CH3+ + CH3OH (CH3)2OH+

(CH3)2OH+ + e- CH3OH + H

DR of (CD3)2OD+

YES !

Production of (CD3)2O only 6 %) !

Grain surface process for formation of dimethyl ether unlikely(Ehrenfreund and co-workers, 2006)

AND:

Anions in space ?

- Negative charge allegedly mostly in form of electrons

- Some anions (OH-, CN-, C- and CH -) found in Halley’s coma (Chaizy et al. 1991)

- CNO- and possibly HCOO- in interstellar ices (Pontopiddan et al. 2002, Schutte et al. 2001)

- anions and cluster anions present in Earth’s ionosphere

Halley 1986

Possible importance of anions in space ?

- Role of atomic anions in early universe

H+ + H- H + H

- Diffuse interstellar bands: possibly PAH anions and carbon-chain anions

- CNO- and possibly HCOO- in interstellar ices (Pontopiddan et al. 2002, Schutte et al. 2001)

- High electron sticking coefficient of lage PAHs

- Anion abundance constrains electron density

Anion chemistry in space

Photodetachment

AB- + h AB + e-

Mutual neutralisation

AB- + C+ AB + C other neutral products

very little experimental data

”The negative charges may reside more in the form of anions than electrons and mutual neutralization may replace dissociative recombination as the main mechanism for removing positive ions.”Alex Dalgarno, 1999

DESIREE storage ring

Double Electrostatic Ion Ring Experiment

Features of DESIREE

Cryostat cooling down to at least 10K

No restriction on ion mass

Electrospray ion source for large ions (PAHs)

Windows for laser spectroscopy