Interstellar chemistry - ACS Publications

Interstellar Chemistry. R. Carbo1 *and A. Ginebreda. Seccio de Quimica Quantica, Department of Chemometrics, Instrtut Quimic de Sarria, Barcelona-17, ...
0 downloads 0 Views 4MB Size
Interstellar Chemistry R. Carb6'

and A. Ginebreda Seccio de Quimica Quantica, Department of Chemometrics, lnstitut Quimic de Sarria. Barcelona-17, Spain Although the discovery of the first interstellar molecular species comes from the early '40's with the detection by optical means of CH, CHt, and CN ( 1 3 ) ,the true birth of interstellar chemistry can he placed in 1963 when the OH radical was identified usine radiotelesco~etechniuues (4). The enormous technological development cxprrienced by radioastronomy durine-the vast . 20 wars has heen reflected in more than 90 new molecular species (including isotopic varieties) uneuuivocallv identified in the interstellar medium (Table 1). and this number increases every year. Parallel to the experimental radioastronomic research, a considerable amount of theoretical work has been done, and a t the present time astrochemistry can he thus properly considered as a new branch of chemistry with its own status as an independent discipline. In spite of its youth, interstellar chemistry has made suhstantial contributions to the better understanding of different phenomena of both chemistry and astronomy. Its importance can be evaluated by considering the relevance of some of the ouestions with which it is concerned: the distrihution of matter in the Galaxy, the origin of stars and the solar system itself (interstellar molecular clouds are thought to be sitesof star birth), the determination of thecosmic isotopic ratiusof the diverse elements and their direct relation to theorigin of the universe, the implication of interstellar molecules in the oriein of life. etc. are several of the ~roblemsin whirh interstehar chemistry is directly involved. Althoueh the suhiect of the resent paper has been extensively revkwed by s&eral authors (5-i9j,we have found no Our aim is thus to references to the tor~icinTHIS JOUHNAI.. expose in a concise-way some of the features which characterize the present status of interstellar chemistry. As the field is too broad to be covered within the space limitations of this article, we have focused our attention to those aspects more strictly related to formation and destruction of molecules, which we feel are more relevant to the chemistry community.

~. ~

~

The Interstellar Medium The matter of our Galaxy-the Milky Way-is distributed between stars (-90%) and the so-called interstellar medium (-10%). The interstellar material, mainly hydrogen, the most abundant element of the universe (see Table 2), comes from either original (nonprocessed) material or the residues of the nuclear hurnine of former stars. The chemical elemental composition of the intersellar medium is currently supposed to ht: similar to that of the entire cosmos (see Table 21. As can be inferred from observational data, the distribution of interstellar mass is not uniform: instead, most of space is almost empty, with a typical density of 4 . 1 &icles/cm3, and matter accumulates in "clouds." The cloud material is not homogeneous, and it is composed of a solid fraction or "cosmic dust" of uncertain (and controversial) (20,21)romposition (-1%) and gas (-99%). Traditionally two kinds of interstellar clouds are distinguished, according to their physical properties and chemical constitution. The first type is the "diffuse" cloud characterized by low density and high temperature (see Table 3), which is easily penetrated by the starlight, and this precludes the

-

' Author to whom correspondence should be addressed. 832

Journal of Chemical Education

formation of complex molecules. Consequently, these clouds' chemistry is reduced to that involving atomic or simple Table 1. Interstellar Molecules (14, 36,37) Molecule H* CH+ CH OH CI CN CO NO

CS SiO SO NS SiS Hz0 c2H

HCN HNC HCO HCO+ N2Hf

H& OCS

so2 HNO

HCS+ NHJ

c&

Detectiona

Molecule

DBtBctiona radio radio radio radio radio radio radio radio radio radio radio radio radio radio radio radio radio radio radio radio radio radio radio radio radio radio radio

HNCO

UV,iR

VIS VIS.radi0

H2CS GIN HNCS

radio UVJR

CHID CH2NH CH2C0 NH2CN HCOOH C4H HCIN CH30H CHICN NHSHO

VIS.radi0 UV.IR.radio radio radio radio radio radio radio radio radio radio radio radio radio radio radio radio radio radio radio radio

CHISH CWW CHlC2H CH3CH0 CeHsCN HCsN HCOOCH3 C W S N C2H50H CH30CHs C2HsCN HCIN HCsN

IR

radio 'UV: ulmviolst:VIS: visible: IR: infrared: radio: rniuowaue w radiohsguancy. mi=immification -ins mnbwersiai. W O

Table 2. Cosmlc Abundances of Elements, Relative to Hydrogen 17)

Element H He

o

C N NB

Fe

si

~g

Abundance

Element

1.00 0.09

S

Abundance 2X 6X

Ar AI

7 x lo-* 3 X lo-* 9 X 10+ 8X 4 x lo4 3x 3 x 10-5

2x 2X 2X 2X 7x

Ca Ni

Na Cr

10-6

lo-' lo-e 10-

lo-' 4 x 10-7 3 x 10-7

ci P

Table 3. Inlerstellar MaHer: Dlarlbutlon and General Prooertles * TDtal mass of interstellar matter: 5 X l o g M a b Gas: - 9 9 % Dust: -1 % Density Fraction by Temperature Region (OK) (particlec m 3 ) Mass (%)

lntercioud Diffuse clouds D B , ~clouds

87000 100 10

'/\daptod hrm refs. (8)and ( i2).

S0.2 1 lo2 lo2 - 10'

'1 M@111olermas) = 1.989 X 1 P g.

-

20 40 40

Fraction by Volume (%) 90 -10 4.5

snecies. On the other hand. the second t. w.e is the "dark" cloud which is denser, colder, and opaque to starlight, circumstances that allow the existence of more complex chemical species. In fact, dark or dense clouds have been the main source of the majority of the interstellar molecules discovered. The physical conditions of the interstellar medium are so unusual. if cornoared to those available a t present in the terrestrial laboratories, that they permit the &-going existence of some unstable and reactive (from our "terrestrial" point of view) species such as molecular ions (CH+, HCOf, N2Hf, etc.) or radicals (OH, HC4, etc.). In this sense, the interstellar medium must he regarded as a unique lahoratory. lnterslellar Chemistry

formed to the eas ohase. While steps (1) and (2) are supposed to occur with &s;)nahle efficient; (i.e., hydrogen at&s are exoected to collide with interstellar grains with a 3U-10070 extent of reaction), the desorption stepby means of the energy liberated during the exothermic reaction seems possible only for molecular hydrogen. For other, heavier molecules, additional desorption mechanisms must be invoked; photodesorption, thermal evaporation, and collisional ejection are some of them. In summary, model building and numerical predictions based upon grain-catalyzed chemistry are a t the moment tentative (with the probable exception of molecular hydrogen) and subject to uncertainty. For that reason, much more attention has been directed to gas-phase chemistry.

General

Gas-Phase Chemistrv

The growing catalog of interstellar chemicals has caused astrochemists to look for qttalilative and quamtitative models which would enable them u)explain how molerules are tinned and destroyed in the interstrlliu environment. The drastically diiferent physical conditions found in thin medium cnntrast severelv with those available on the Earth, and, accordinelv, .. a suhs&tially different chemistry must he expected. Since the first studies undertaken by Spitzer and Bates (22) concerning CH and CHf, much work has been done and some of the questions clarified. Speaking in general terms, it can he said that the qualitative picture of interstellar chemistry is reasonably clear, while its quantitative aspects remain uncertain. Due to the very low temperatures and densities prevailing in the interstellar medium, its chemistry becomes restricted to the following process: (a) exothermic reactions; (h) himolecular reactions (three-body collisions are very improbable) ( 9 )

A qualitative list of the reaction types that are supposed to he involved in the gas-phase chemistry of the interstellar medium is given in Table 4. Reactions between positive ions and neutral species are currently thought to be the most likely processes; their superior efficacy if compared to neutralneutral reactions arises because of their lower activation energies. In order to discuss more properly the gas-phase chemistry of interstellar molecules i t i s convenient to treat separately diffuse and dark clouds, as their chemistry is controlled by entirely different processes. Diffuse Clouds Diffuse clouds are more or less penetrated by the galactic diffuse starlight, which is predominately in the 1000-2000 A wavelength range. Such ultraviolet photons are the principal

Table 4. Gas-Phase Reaction Types ot Importance In Interstellar Chemistry In addition. exothermic reactions are likelv . onlv.if thev have a reasonably high rate constant r, which results if the activation enerw E,- of the Arrhenius eauation is comparable to the thermic energy kT

--

A++%-Bt+A b) Radiative association

A++%-AB++h" C)

= A . e-EdkT

A t m transfer

Af+BC-ABf+C

Two types of mechanisms seem to fit the previous reauirements: -eas-phase chemistrv hased uoon . . .oositive ionmolecule reactions and formation of molecules on the surface of rmmic dust rrains. We will consider both schrmes seoarately. Some other alternative processes such as synthesis in stellar atmospheres followed by ejection of molecules during stellar evolution or synthesis in shocked clouds, which sometimes are invoked, will be omitted in the present discussion. The reader is referred to the original papers such as (23). Grain-Catalyzed Chemistry

AB++C-A++BC

11. Electron recombination reactions a) Radiative A++e--A+hu b) Dissociative

ABf Ill. Photochemical reactions a) Photcdis~oCiation

+

.--

A

+B

AB++~v-A++B b) PhOtOionizatiOn

Due to the large uncertainties that still remain about the physical and chemical constitution of cosmic dust, it is difficult to make a realistic evaluation of the importance of grain-catalyzed surface reactions to the synthesis of interstellar molecules. Nevertheless. i t seems necessarv to consider such a orocess to account for the observed ahundances of molecula~hydro-een.. which cannot be exnlained exclusivelv. hv- eas-phase - . chemistry. H + H + grain

I. Ion-molecule reactions a) Charge transfer

-

Hz + grain

(1)

The sequence of steps needed t o produce a molecule hy surface chemistrv are the followina: (1) adsorption of atoms on the dust grainburface, (2) an encounter hetween the atoms involved in the reaction, and (3) ejection of the molecule

~ + h v - ~ + + e IV. Neutral-neutral reactions a) Atom (or atom group) transfer

A+BC-AB+C b) Radiative association

A+BjAB+h" C)

Chemionization A+B-AB++B-

V. Other reactions 8) Ion-ion neutralization

A++B-b) Negative ion-neutral

AB

A-+B-AB+eA-+%C-AB-+C

Volume 62 Number 10 October 1985

833

external enerav source. and thev. eovern the ionization and .. destruction p&essrs (,f atomsimd moleculeswithin the cloud. The nhsurption ointotnir hydrogen cuts off the spectrum of . (13.6 the light w photons with u,a\.elengths greater than 912 & ~ Y Iwhich , rorrt:s~ondato hvdn~ren'sionization notentinl. ~ h energetic h harrier divides the elements in t w i groups: those whose ionization potential is lower than 13.6 eV (such as C Jexist mainl) as ions, while those with ionization potenti313 greater than this value rrrnnin in neutral form (i.e., 0, NJ. The ihemical composition of diffuse clouds is thus very simple and the only availahle molecules are mostly diatomics snch as HZ,HD, OH, CO, CH, CH+, Nz, NO, CN, 02, etc. The accepted formation mechanisms for some of these species are given in Table 5. I t must be noticed that the CH and CH+ chemistry, whose study began more than 30 years ago, still remains controversial and one of the unsolved problems of interstellar chemistry. Dense Clouds In these clouds, opaque to ultraviolet starlight, photodissociation and photoionization are negligible, and they have a unique external energetic source-the cosmic rays (i.e., highly energetic nuclei of 1-100 MeVInncleon), which are present throughout the whole Galaxy. Cosmic rays are responsible of the initial ionization step of hydrogen and helium, the two preponderant elements Cosmic ray Cosmic ray

+ Hz + He

--

Hz+ + eHe+ + e-

(2)

The next step is the reaction between these ions and other species to produce highly reactive ions, the most important being Ha+, formed by reaction of Hz+ with molecular hydrogen Hz++Hz-Hs++H

H3++A-AHC+Hz A = CO,Nz,O,N,Cz AH+ = HCO+,N2H+,NHC,C2H+,0HC

The detection of HCO+ and NzH+ provides good support for these mechanisms. In turn, helium ions can dissociate molecules producing atomic ions according to the following reaction

Reaction (5) is the main source of atomic ions, as other processes, snch as photoionization, are unimportant in dense clouds. Ionic molecular species may react with molecular hydrogen (the most abundant chemical in dark clouds), giving rise t o ions increasingly hydrogenated A++H2--AH++H

Those ions that do not react further with molecular hydrogen are destroyed by dissociative electron recombination. Some of the known interstellar species (i.e., OH, H20, CH) are produced in this way

Proposed Formation Mechanisms for Some Selected Species in Diffuse Interstellar Clouds

Ammonia is formed by charge exchange between NH3+ and

Table 6.

Proposed Formation Mechanisms tor Some

Intermediate Complex Molecules in Dense Interstellar Clouds --. -- -. -CN.HCN.HNC NH3 C+

+

H&NC

H2C0 a) CHI

+ e-

+0

+ Hz

b) HCOf

H&OC

c) CHaC H&O+

+ e-

+0

+M

M

Fe,Mg.Na

834

Journal of Chemical Education

(4)

(3)

The H3+ ion tends to react with molecules through reactions

Table 5.

of the type shown in eqn. (41, yielding hydrogenated molecular ions

. . . etc.

---

+H + H2 HCN + H H&N+ HCNt

---

+H + hv +H H.CO + H H&O+ + H M+ + HzCO H&O

H&O+ -, H&O+

metals of low ionization potential (i.e., Na, Mg, Ca, Fe) or by dissociative electron recombination of NH.+. The syntheti(. schmws for some molecules of intermedintc comolexitv such as H(:N. HNC. HKO. nnd CU are recorded in r able 6 The key problem of interstellar chemistry is how the more complex organic molecules are formed. Much effort has been spent in the field, including both the experimental determination of many rate constants of gas-phase reactions and theoretical calculations; nevertheless, many questions remain unsolved. An exhaustive exposition of all the work in this area is heyond the scope of this review, but some indications about the most promising trends of gas-phase chemistry of complex molecules can be outlined. Chemistry of Cyanopolyynes. According to Winnewisser et al. (24), the carbon chain of cyanopolyynes can be built by a successive incorporation of acetylene units through the following reaction The terminal cyano group is incorporated either by reaction with CN or HCN, which are hoth common interstellar species

+

H2C.+ CN H2C,+ HCN H&+ CN H&+ HCN

+

+

+

---

+ + +

HCnclNC H H2C,+~NC H H2C,+~N+ H H2C,+,Nf + HZ

of simple linear molecules, a circumstance that makes difficult their radioastronomical detection. Synthesis of Complex Molecules by Radiative Association Reactions. Radiative associations of the type A++B-ABfthv

followed by electron recombination (usually with abstraction of an hydrogen atom) have been recently reexamined by several authors (25-28), and it seems that they can account for the formation of many of the more complex molecules. As an example, the proposed interstellar synthesis of methanol is given

+

CH30H2+ e-

+ CHJOH + H

CH80H2+ hu

A+B+M-AB+M

(16)

A+B+AB* AB*+M-AB+M

(17)

or

The last step is substituted in alow-density medium, such as interstellar clouds, hy radiative stabilization AB*-AB+hv

If such a mechanism is correct, substituted polyynes with groups different from cyano (i.e., C H O , -NH2, -OH) should be abundant too. These molecules have not yet been detected, probably because they are asymmetric rotors instead

Table 7. Radlatlve Assoclatlon Formatlon Mechanlsms Proposed for Some Complex Molecules in Dense Interstellar Clouds, accordina to ref. (25) General mechanism: A+ + X

-

A Xf

+ hv

AX++e--B+HorH2 A Xf:

Parent Ion

Neutral

(A+)

(XI

CH,+

hydrogenated ion Final Molecule (0)

HI H20

+ -

(CH.+

CHs+

CH*O+ 0 CH30f

+ hu)

+ HZ)

co HICO HICO

Quantitative Models The eeneral Durpose of quantitative models of interstellar chemish is tomnke prediciiona of the deniitiesof rhe various chemical sr,rcies that aaree umell with those inferred from ubservationai radioastro~omicdata, in order to validate the supporting reaction schemes. Actually, several quantitative models appeared in the current literature (29-35, 38), with variable degrees of sophistication. However, all of them show some features in common

The main differences lay in the following aspects:

HCOf

HI H2O

H&08 HCOOH

CzHs*

H20

CH3CH20HS

NHQf

0 CO HCN H&O

HNOe HNCO NH2CNe NH2CH0

H2

HNO'

NOC

Finally, surface reactions cannot be ignored, although their relative importance in the synthesis of complex organic molecules is a poorly understodd matter.

depth. and rlcmrnr nbundanrr* is asiurnrd for the model rloud. Temperature controls many rate constants, while optical depth (which is interdependent with total density and cloud mass) is mainly concerned with photochemical processes. Element abundances are usually taken equal to the cosmic ones, and they act a s additional constraints. 2) -. A series of chemical reactions (each one characterized bv their rcsprrtivr rntp wn+tant, i i set up, and t he suhiequent system of kinetic er(untiunardred.

H2C0 CH30H NH. HCN

CH$

+ HI

(18)

I , A set ~dbtartingcmdirionsof totaldensity, lemperature,optical

co

(CHa+

(15)

By appropriate variation of the parent ions and the neutral molecules (see Table 7) the majority of the interstellar species can be exdained. ~ a d i a t i v eassociation reactions have been experimentally studied bv com~arisonwith analo~ous - three-bodv association reactions-

(10)

followed by electron recombination

+ -

CH3+ H 2 0

(9) (11) (12)

(14)

AltematiYe processes have been pmtuhlated

1) ~. The number of chemical soecies considered.

2, The number d rrarti~micr~nuidered. :3j

The rsritnorcs for the valuer of the raw ronstants. mathcrnar~rnlresdutam pnwcdurc of the kmetie equation system.

4 ) The

Information relative to points (I),(2), and (4) is summarized in Table 8 for the most important published computations, and they need no further comment, except for the last point (4): two mathematical approaches have been used. On the one Volume 62 Number 10 October 1985

835

Table

8.

A C o r n p a r l s o n cd O u a n t l t a t l v e M o d e l s o f I n t e r s t e l l a r C h e m i s t r y

Number of Soecies

Number of Reactions

Calculation Tv~e

Graendel-Lenger-F~erking(37)

100 137 126

455 1423 1067

steady-state timedependent timedependent

Tielens-Hagen (33)

139

1520

steady-state

PicklesWilliams (34)

...

...

steady-state

Millar (35)

95

21 1

steady-state

wan (38)

89

1427

Mcdei (Ref.! M i t c h e l l G i n r b u r ~ K u n t z(29) Prasad-Huntress (30.31)

hand, there are some models which make use of the steadystate assumption; i.e., if we represent the rate of formation1 destruction for each chemical species x i by xi, it will be expressed as a nonlinear function of all the concentrations x = 1x1. . . x,) (n being the number of species considered): f;(x) We can express the system in a vectorial form as where =

{

=

-1

dri

dt

... ... 12 Grain-surface reactions included. Isotopic Substituted varieties of C a n d 0 included. G r a i n l u r f a c s chemistry included. Grain-surface chemistry included. S o m e reactions treated as parameters. 20 Grain-surface reactions included.

timedependem

...

Douglas. A. E., and Henberg, G., Astraphys. J , 94,381 (19411. Dunhsm,T.,Publ. Aatr Soc.Pac.49.26 (19371. Weinreh. S.. et al.. Nature. 200,829 (19631. Winnewker, C., M ~ ~ p ~ r , P . G . , a n d B r ~D~ e. r, , 'H' ~~ n ~ ~ ~ U ~ M b"Topim o I e c d ~ ~ in Current Chemistry." Springcr~verlag,New York, 1974,Vol. 44, PP. 1-83. (6) Herbst, E.. and Klemperer, W., Phys Today, 29.32 (1978). (71 Gmman,R. H., Chem. & Eng. Nerus.56[401.21(19781. (8) Ta1bnt.R. J.,lnf. Set. Reu..5[21. 102 (18801. (91 Watson. W. D.. Rau. Mod. Phva.. 48141.513 (19761. (21 (3) (4) (5)

and f = Ifi(x)I

The steady-state method computes the stationary concentrations solving the set of equations f = 0. In other words, an equilibrium situation is assumed. On the other hand, there are models that integrate the differential nonlinear system of equations, computing a time-dependent solution for every species. The appropriate question is whether the lifetime of the cloud is long enough to allow the steady-state to be reached or not. It seems that at t = 106-107 y the equilibrium is accomplished for the majority of the species. This is roughly a value of the same order (or perhaps one order less) of that estimated for the cloud lifetime. A second differentiating aspect to be mentioned is grainsurfacechemistry, which has been employed to variable extent on some of the models already published (32-35) (the exception is, of course, the grain formation of Hz, which has been considered throughout all the models). Although many of the quantitative models agree with remarkable coincidence with the experimental m e a s u r e m e n t s , the principal criticism to be madeis the omission of the more complex molecules in these calculations. There is no doubt that increasing computing facilities, together with t h t . vxperimental d e t e r m i n a t i o n of more kinetic constants, will help to improve the results

Acknowledarnent One of us (A.G.) is indebted to the Direcci6 General $Ensenvament Universitari de la Generalitat de Catalunva for a posidoctoral fellowship. Llterature Cited (1) McKellar.A.,Publ. Adr. Soc Poc.,52,187 (19401.

836

Omer Remarks

Journal of Chemical Education

( 1 s ~ u n t ~ e sW.T., s , Chem. Sac. ~eu.;6,3W (19771. (19) Snyder. L. E., and Buhl. D.. Sky & Telescope, 40.1(19701. (20) Lequcur, J., Geochim. Cosmoehim. Aeto.46.777 11982). (211 Savage, 8.D., and Mathis, J. S., Ann. Re". A s t ~ ~An s. t r ~ p h y ~I.S. 7 3 (1978) 1221 Soitzer.L..and B8te.D. R..Astroohvs. J.. 113.441 119511.

,.

s S .and ~ u n t r n .W T , . ~ \ t r o p h > *I Suppl .13. I llr80,. S and H m ~ r r sW. . T . Arrr oh,, J ,219, IS1 11Cd I u 11, A ~ , IF I P ~ ~ , ,M, ~ ..+.. ~ ~ r J S~U P P~I , I, ~ 32, (;raen~d. 'r i..I

I h , i'raad

',I, P r n a L S

h

(19821. (331 Tioions, A. G. G. M., and Hsgen, W., Aatron. Aalrophys., 114.245 (19821. 184) Pickle. J.E..and Williame.0.A.. Man. N0f.R. Asfr Sor.. l97.429i1981J. ~ ,. , (381 Wett,C.D., Mon Not. R. Astr Soe.205.321 (19831,

Note Added in Proof After submitting this D a D e r for ~uhlicationthe followine new interriellar ib~ecuies'havebeen discovered: CH?C,H: CsO. HCIIN, SiCz, and prohably COH'; and the following calculations on kinetic q t l a n t i t n t i v r models published: h1ll~hell.C.F., I o n ' o r . R Arrr So? ,205.7W 19d31 li~.!,~rR,,,.?, ,A,. .I .S,.~,71,63,d, ,,98: