X. Gu et al.
Kinetics of Triacetylene Formation with Application to Titan
X. Gu, Y.S. Kim, R.I. Kaiser*
Department of Chemistry, University of Hawaii at Manoa, Honolulu, HI
A.M. Mebel
Department of Chemistry and Biochemistry, Florida International University, Miami FL
M.C. Liang
Research Center for Environmental Changes, Academia Sinica, Taipei, Taiwan
Graduate Institute of Astronomy, National Central University, Jhongli, Taiwan
Y.L. Yung
Division of Geological and Planetary Sciences, Caltech, Pasadena, CA
Author contributions: RIK designed research; XG, AMM, YLY, and MCL performed research. RIK
wrote this paper.
The authors declare no conflict of interest.
This article is a PNAS direct submission.
* To whom the correspondence should be addressed to. E-mail: ralfk@hawaii.edu
1
X. Gu et al.
Abstract
For the last four decades, the role of polyynes such as diacetylene (HCCCCH) and triacetylene
(HCCCCCCH) in the chemical evolution of the atmosphere of Saturn’s moon Titan has been a
subject of vigorous research. Polyacetylenes are thought to serve as an ultraviolet radiation shield in
planetary environments thus acting as prebiotic ozone and present the key intermediates in the
synthesis of more complex polyynes such as tetraacetylene (HCCCCCCCCH) leading ultimately to
the visible, organic haze layers on Titan, which are thought to resemble terrestrial aerosol layers
before life developed on Earth more than 3.8 billion years ago. Nevertheless, the underlying
chemical processes which initiate and control the haze formation have been the least understood to
date. Here, we present a combined experimental, theoretical, and modeling study on the synthesis of
the triacetylene molecule (HCCCCCCH) via the bimolecular gas phase reaction of the ethynyl
radical (CCH) with diacetylene (HCCCCH). This elementary reaction is rapid, has no entrance
barrier, and yields the triacetylene molecule via indirect scattering dynamics through complex
formation as the sole reaction product. This presents the very first experimental detection of the
triacetylene reaction product under single collision conditions as prevailing in Titan’s atmosphere in
those regions where density profiles of photolytically generated ethynyl radicals and diacetylene
molecules overlap. New photochemical models imply that this molecule may serve as a key building
block to form more complex polyynes in Titan’s atmosphere leading ultimately to the organic aerosol particles, which make up the orange-brownish haze layers in Titan’s atmosphere.
2
X. Gu et al.
Introduction
The arrival of the Cassini-Huygens probe at Saturn’s moon Titan - the only Solar System body
besides Earth and Venus with a solid surface and thick atmosphere – in 2004 opened up a new
chapter in the history of Solar System exploration (1). The mission revealed Titan as a world with
striking Earth-like landscapes involving hydrocarbon lakes and seas as well as diverse fields of
organic sand dunes interspersed with craters and icy mountains (2). The discovery of a dynamic
atmosphere and active weather system (3-6) illustrate further the similarities between Titan (7) and
Better to replace ref 3 by papers by Held, Holton, Lindzen, and etc, such as
Baldwin MP, Gray LJ, Dunkerton TJ, Hamilton K, Haynes PH, Randel WJ, Holton JR,
Alexander MJ, Hirota I, Horinouchi T, et al. (2001) Reviews of Geophysics 39, 179-229.
Yuk should have better suggestions
Better to include papers by Rannou et al. such as
Rannou P, Hourdin F, & McKay CP (2002) Nature 418, 853-856.
This is the right format for PNAS.
Earth. The aerosol-based haze layers, which give Titan its orange-brownish color, are not only
Titan’s most prominent optically visible features, but also play a crucial role in determining Titan’s
thermal structure and chemistry (8). These smog-like haze layers are thought to be very similar to
those present in Earth’s atmosphere before life developed more than 3.8 billion years ago (9,10) and
absorb the destructive ultraviolet radiation from the Sun thus acting as ‘prebiotic ozone’ to preserve
astrobiologically important molecules on Titan (11-18). Compared to Earth, however, Titan’s low
surface temperature of 94 K and the absence of liquid water preclude the evolution of biological
chemistry as we know it. However, exactly because of these low temperatures, Titan provides us
with a unique prebiotic “atmospheric laboratory” yielding vital clues – at the frozen stage – on the
chemical composition of the atmosphere of primitive Earth a few billion years ago (19) (not so
applicable, temperature a few Myr ago remain much higher than Titan’s; also $$$ proto-Earth has a
different meanin $$$
), although primitive Earth might have had a higher oxygen content than Titan (20).
However, despite the importance of Titan’s organic haze layers for understanding the origin and
chemical evolution of hydrocarbon-rich atmospheres and of primitive Earth, the underlying
3
X. Gu et al.
chemical processes, which initiate and control the haze formation, have been the least understood to
date (21).Based on limited laboratory studies related to the formation of soot ($$$ Allen et al paper
is in http://yly-mac.gps.caltech.edu/ReprintsYLY/N035… $$$
, planetary chemists proposed that ethynyl radicals, (C2H (X2+)), which are generated by photolysis
of acetylene (C2H2) at wavelength less than 217 nm, react with unsaturated hydrocarbons via
organic transient species to form haze (22,23). These considerations have led to the development of
photochemical models of Titan (24-29) and also to extensive laboratory studies during the last
decades (30-33). It is remarkable that, based on reasonable alternative choices for the unknown
reaction dynamics, photochemistry, and reaction rates, the models show completely inconsistent
mechanisms for the principal routes to hydrocarbon build-up in Titan’s aerosols (34). Several limitations of the experiments such as wall effects undermine their validity (35). Also, the reaction
products are either guessed or often analyzed off-line and ex situ (36-38). Hence, the detailed
chemical dynamics of the reactions such as the role of radicals and unstable, transient species cannot
always be obtained, and reaction mechanisms can at best be inferred indirectly and qualitatively.
Since the macroscopic alteration of Titan’s atmosphere consists of multiple elementary reactions
that are a series of bimolecular encounters between radicals and molecules, a detailed understanding
of the underlying mechanisms involved at the microscopic level is crucial for unravelling the
chemical evolution and processing of low temperature environments in general. These are experiments under single collision conditions in which particles of one supersonic beam are made to
‘collide’ only with particles of a second beam utilizing crossed beams experiments.
Data from the Cassini-Huygens mission have revealed that the transformation of simple
molecules like acetylene (HCCH) and diacetylene (HCCCCH) to complex species such as aerosols
presents the most fundamental step in the context of planetary evolution ( 39 ). Therefore, an
experimental investigation of these elementary reactions under single collision conditions is clearly
crucial (40,41,42). In this article, we report the results of the crossed molecular beam reactions of
the isotope variant of the elementary reaction of ethynyl radicals (C2H; X2+) with diacetylene
(C4H2; X1g+). These results are combined with electronic structure calculations and photochemical
modeling to unravel the formation and role of the organic transient molecule triacetylene (C6H2;
X1g+) in the evolution of the organic haze layer in Titan via bimolecular collisions between two
neutral species (equation (1)).
4
X. Gu et al.
(1)
C2D(X2+) + H-CC-CC-H(X1g+)  D-CC-CC-CC-H(X1+) + H.
This system represents the prototype reaction of ubiquitous atmospheric ethynyl radicals in Titan
with the simplest polyacetylene species, diacetylene, to form triacetylene in hydrocarbon rich atmospheres of planets and their moons via a single collision event. Both the diacetylene precursor and the
triacetylene product have been suggested to be central building blocks (34) in the complexation of
the organic aerosol layers in Titan’s atmosphere; consequently, a coupling of the laboratory experiments with electronic structure calculations and photochemical reaction networks will provide key
information and elucidate the role of these polyacetylenes in the chemical evolution of hydrocarbon
rich atmospheres. This presents the missing link to untangle the formation of the aerosol layers from
the ‘bottom’ up via reactions of simple radicals with hydrocarbons such as complex polyacetylenes.
Results and Discussion
In our crossed beam experiments, we monitored reactive scattering signal at mass-to-charge
ratios of m/z = 75 (C6HD+) and 74 (C6D+/C6H2+) (Fig. 1). Data at m/z = 75 could be fit with a single
reaction channel of the mass combination of 75 amu (C6HD) and 1 amu (H). This finding alone
verifies the formation of a molecule of the gross formula C6HD under single collision conditions.
Signal at m/z = 74 can in principle arise from three contributions: the ionized, heavy product of the
C6D (74 amu) + H2 (2 amu) and/or C6H2 (74 amu) + D (2 amu) channels and/or dissociative
ionization of the C6HD reaction product (75 amu) in the electron impact ionizer of the detector. The
experimental results depicted clearly that the TOF spectra of m/z = 75 and 74 are identical; this
suggests that signal at m/z = 74 results only from dissociative ionization of the C6HD neutral (75
amu) in the ionizer of the detector; if either the C6D + H2 or channel C6H2 + D channel is involved,
the shape of the TOF spectra should be different. Therefore, both the molecular hydrogen elimination channel and the atomic deuterium exchange pathways are closed. To summarize, the analysis of
the TOF spectra alone indicates that the formation of a molecule of the formula C6HD plus a light
hydrogen atom from the former diacetylene molecule is the only open pathway in the crossed beams
reaction of the D1-ethynyl radical with diacetylene.
We also investigated the chemical dynamics of the reaction as this assists unraveling the
nature of the C6HD product isomer and the intermediates formed. Recall that these hydrocarbon
species have a strong influence on the planetary chemistry and the hydrocarbon balance. The infor-
5
X. Gu et al.
mation on the chemical dynamics are extracted by transforming the laboratory data (TOF spectra
and laboratory angular distribution; Figs. 1 and 2, respectively) into the center-of-mass (CM)
coordinate system to yield a center-of-mass product flux contour map (43) (Fig. 3). The plot reports
the differential cross section, I(,u), as a function of product center-of-mass scattering angle  and
velocity u; the flux contour plot is proportional to the product of the center-of-mass angular distribution, T(), and the center-of-mass velocity distribution, P(u), which is turn is computed from the
center-of-mass translational energy distribution, P(ET). The flux contour map, I(,u), can be seen as
an image of the reaction and holds all the information of the reactive scattering process. First, an
inspection of the center-of-mass translational energy distribution allows computing the reaction
energy and hence helps to elucidate the nature of the C6HD isomer formed. The maximum translational energy, Emax, released into the products represents the sum of the collision energy, Ec, and
the absolute of the reaction energy, rG. Therefore, by subtracting the collision energy (Ec = 40.2 ±
2.0 kJmol-1) from the maximum amount of translational energy (Emax = 160 ± 14 kJmol-1), we can
compute the reaction exothermicity to be 120 ± 16 kJmol-1. This agrees very well with our ab initio
data of 125 ± 5 kJmol-1 to form the D1-triacetylene isomer (HCCCCCCD) plus atomic hydrogen
(H). Secondly, the translational energy distribution peaks away from zero translational energy at
about 35 ± 5 kJmol-1. This finding indicates that the exit transition state is likely to be tight and
involves a significant electron rearrangement from the decomposing intermediate to the final D1triacetylene plus hydrogen atom products; in other words, the reversed reaction of a hydrogen atom
addition to D1-triacetylene involves an entrance barrier. Thirdly, the center-of-mass angular distribution, T(), shows intensity over the complete angular range from 00 to 1800. This pattern is
characteristic of an indirect reaction mechanism, i.e. the formation of a C6H2D reaction intermediate
which decomposes via atomic hydrogen loss to form D1-triacetylene. In addition, the center-of-mass
flux contour plot is slightly forward scattered with respect to the D1-ethynyl beam indicating that
the lifetime of the reaction intermediate is less than its rotational period (44). In a molecular beams
experiment, the rotational period can be used as a clock to estimate the lifetime of the decomposing
complex. Based on the intensity of the poles of I(1800)/I(00) = 2.5
fragmenting C6H2D intermediate can be approximated to be 0.55
0.13
 0.17
 0 .5
 0 .8
, the lifetime of the
of its rotational period (45).
To elucidate the underlying chemical dynamics and the reaction mechanism, we are merging
now our experimental findings with our computational study of the reaction of D1-ethynyl (CCD)
6
X. Gu et al.
with diacetylene (HCCCCH) (Fig. 4). Most important, our computations verify the experimental
results of an indirect reaction mechanism involving C6H2D reaction intermediate(s). Here, the
reaction is initiated by a barrier-less addition of the D1-ethynyl radical with its radical center to the
C1 and/or C2 carbon atom of the diacetylene molecule yielding intermediates [1] and [3],
respectively. These structures are bound by 294 kJmol-1 and 226 kJmol-1 with respect to the
separated reactants. Intermediate [3] connects [1] and [2] via a D1-ethynyl group shift from the
diacetylenic C2 to the C1 carbon atom. The calculations identify the C6H2D structure [1] as the
decomposing intermediate which emits atomic hydrogen. The tight exit transition state (found
computationally), which ranges about 21 kJmol-1 above the separated D1-triacetylene and atomic
hydrogen products, was predicted experimentally based on the off-zero peaking of the center-ofmass translational energy distribution (Fig. 3). A comparison of the geometries of [1] and the final
triacetylene product verifies the electronic reorganization involved in this process and hence the
expected tight exit transition state. Finally, the overall reaction was computed to be exoergic by 125
± 5 kJmol-1. This data agrees exceptionally well with the experimentally derived value of 120 ± 16
kJmol-1 to form D1-triacetylene plus atomic hydrogen. It should be stressed that the second lowest
energy isomer of C6H2, i.e. the carbene structure CCCCCCH2, is at least 210 kJmol-1 less stable than
triacetylene ( 46 ) and, hence, can neither be formed in our experiments nor under conditions
prevailing in Titan’s atmosphere.
We are transferring now our findings from the laboratory to the ‘real’ atmosphere of Saturn’s
moon Titan and discuss the effect of the title reaction to Titan’s atmospheric chemistry, aerosol
formation, and hydrocarbon balance. First, the results reveal unambiguously that under single
collision conditions, the exoergic reaction of ethynyl radicals with diacetylene can form triacetylene
in Titan’s atmosphere. As elucidated computationally, this process has no entrance barrier. The
absence of an entrance barrier presents a crucial prerequisite for a chemical reaction to be feasible
under the extreme low temperature conditions in Titan’s atmosphere (94 K to 200 K). These low
temperatures typically block reactions with either have a significant entrance barrier or which are
endoergic; therefore, chemically relevant reactions must be exoergic, proceed without entrance
barrier, and must only involve transition states which are lower than the energy of the separated
reactants. All these criteria are fulfilled in the bimolecular reaction of ethynyl radicals with
diacetylene thus expanding the carbon skeleton by one dicarbon unit to synthesize triacetylene plus
7
X. Gu et al.
atomic hydrogen in a single collision event. Secondly, our studies suggest that the triacetylene plus
atomic hydrogen channel is the only pathway relevant to Titan’s chemistry (47). The competing
direct hydrogen abstraction pathway to form acetylene (C2H2) plus the 1,3-butadiynyl radical (C4H)
involves a barrier of about 38 kJmol-1. Considering the low temperatures in Titan’s atmosphere of
94 K to about 200 K, the competing hydrogen abstraction pathway is clearly irrelevant, and
triacetylene plus atomic hydrogen are the sole reaction products. This is very first time that the
reaction pathway to form triacetylene in Titan’s has been identified experimentally and computationally. In addition to the explicit identification of the triacetylene reaction product, our investigation
identified three radical intermediates [1], [2], and [3]. Under our single collision conditions, these
species are formed with extremely high internal excitation, and their lifetimes are less than a few
picoseconds. Hence, in our experiment intermediate [1] fragments before it reaches the detector.
However, in denser atmospheres of Titan, we have to analyze to what extent ternary collisions
impact the chemistry. These processes can divert the internal energy and hence will stabilize those
intermediates if the lifetime of the intermediate is longer than the time to allow a ternary collision to
take place. Based on the potential energy surface (Fig. 4), we investigated the unimolecular decomposition of intermediate [1] utilizing RRKM (Rice-Ramsperger-Kassel-Marcus) calculations. This
treatment suggests lifetimes of intermediate [1] of 3 to 24 ns; these times cover the temperature
window of 94 K to 200 K relevant to Titan’s atmosphere. Taking the atmospherically relevant
pressure window of 1 to 1×10-6 mbar (48), and the collision cross section between intermediate [1]
and a nitrogen bath molecule in Titan’s atmosphere from our computations (1.97×10-18 m2), we
estimate time scales between the collisions of about 17 – 22 ns at 1 mbar and 17 – 22 ms at 1×10-6
mbar. A close look at these time scales indicates that the time between the collisions of intermediate
[1] and a nitrogen bath molecule compared to the lifetime of intermediate [1] of typically a few ns is
clearly too long to allow a third-body collision to be relevant in Titan’s atmosphere in those regions
where the ethynyl and diacetylene reactants are present. Consequently, the intermediates formed in
the reaction of ethynyl radicals with diacetylene cannot be stabilized by third body collisions in the
temperature range of 94 – 200 K and pressure window of 1 to 1×10-6 mbar; therefore, in Titan’s
atmosphere, all intermediates [1] decompose solely to triacetylene plus atomic hydrogen.
Having identified the triacetylene molecule as the sole reaction product of the bimolecular
reaction of ubiquitous ethynyl radicals with diacetylene in Titan’s atmosphere, it is of crucial
8
X. Gu et al.
importance to investigate now the effects of triacetylene on the global chemistry in Titan’s
atmosphere, specifically the fate of the triacetylene molecule and its role in the formation of the
organic haze layers. This transfers our laboratory findings obtained on the microscopic level to the
macroscopic world. Therefore, we modeled the significance of this reaction and implemented the
reaction of ethynyl radicals with diacetylene into a new photochemical model of Titan’s atmosphere.
This process requires two accurate input parameters: the correct reaction products, i.e. triacetylene
plus atomic hydrogen as elucidated in the combined crossed beams and computational study, and
accurate rate constants. Whereas crossed beam reactions provide unique information on the reaction
products, they hardly supply accurate rate constants; also, a detailed literature research failed to
provide any rate constants for the ethynyl – diacetylene system. Due to the absence of any
experimental data on the rate constants, we incorporated for the first time precisely computed rate
constants of 1.5 ± 0.5 ×10-10 cm3s-1 into the model (49). Over the temperature range of 94 K to 200
K, the rate constants are – within the error limits – temperature independent. This finding is characteristic of an exoergic reaction which has no entrance barrier and which is dictated by attractive,
long range forces (44). It should be noted that the computed rate constants hold a similar value as
those experimentally determined for the related reaction of ethynyl radicals with acetylene yielding
diacetylene plus atomic hydrogen (50).
Our photochemical models suggest that polyynes of the generic formula C2nH2, (n  2) are
essential building blocks for the formation of higher hydrocarbon compounds in the atmospheres of
outer solar planets and their moons in general. The richness of the hydrocarbon chemistry is
determined by the relative concentration of methane to hydrogen. For hydrogen-rich Jovian
atmospheres such as Jupiter, hydrocarbon radicals such as C2H preferentially react with molecular
hydrogen. There are two effects for this preferred pathway. First frequent reactions with hydrogen
recycle hydrocarbons back to methane. Second, the decomposition of methane is limited by the
magnitude of direct UV dissociation, whereas on Titan these radicals are efficiently used for
methane dissociation, further enhancing hydrocarbon chemistry and hydrocarbon production.
Consequently the “net” methane loss rate for Titan is a factor of 2 higher than that for Jupiter, even
though Titan receives solar UV photons a factor of 4 less than Jupiter does. Photochemical models
predict relative column densities of [C4H2]/[C2H2], [C6H2]/ [C2H2], and [C8H2]/ [C2H2] to be about
710-5, 110-7, and 110-9, respectively, for the atmosphere of Jupiter. For Titan, these values are
about 10-2 – at least three orders of magnitude higher than those in Jupiter. The modeled profiles of
9
X. Gu et al.
acetylene and polyynes among them triacetylene are shown in Figure 5; such reference model can
well simulate the profiles of various hydrocarbons in the neutral region of the atmosphere observed
by Cassini-Huygens (pls add ref.
Liang MC, Henze D, Adamkovics M, Chu EF, Boering K, Yung YL (2009) Synergistic Study of
Hydrocarbon Photochemistry in the Laboratory and Planetary Atmospheres. Submitted to
Planetary and Space Science
Their productions are significant throughout the whole atmosphere, because of the photolysis of
higher order hydrocarbons that mediates the decomposition of methane. Consequently their
abundances in the lower atmosphere remain high.
We next evaluate how the new reactions alter the aerosol formation. Under the currently observed
temperature profiles in the atmosphere of Titan above the tropopause at about 50 km and the
thermodynamic constants for polyynes, these molecules are not able to condense to form aerosols.
To account for the observed aerosol loading in the mesosphere and thermosphere of Titan, physical
adsorption was proposed (51) and expanded to account for the triacetylene production as elucidated
via the crossed beams experiments and computationally. With an adsorption coefficient of 1%, the
adsorption rates of acetylene (C2H2), diacetylene (C4H2), triacetylene (C6H2), and tetraacetylene
(C8H2) are about 2109, 1108, 3107, and 1107 molecules cm-2 s-1, respectively. As a result of the
adsorption in the higher atmosphere, the deposition rates (the fluxes into the troposphere) of these
species at and below the tropopause are reduced by up to four orders of magnitude. Subsequent
chemical reactions driven by solar UV photons and high-energy cosmic-ray could convert the
molecule-aerosol complexes formed in the atmosphere into refractory organic polymers (tholins)
and then sequester them in the surface (51). Notice that the photolysis of triacetylene at wavelengths
less than 200 and 230 nm leads to decomposition to C4H + C2H (15%) and C6H + H (85%). The two
processes can then catalyze the decomposition of methane and hence mediate more efficient
hydrocarbon production than methane photolysis alone, which requires energetic solar UV photons
with wavelengths less than about 130 nm. Consequently the total decomposition rate of methane is
about 1010 molecules cm-2 s-1; only one third is contributed by the direct photolysis. In contrast, the
major constituent of giant planets such as Jupiter is molecular hydrogen (H2). Hydrocarbon radicals
react with molecular hydrogen more preferentially than methane (52). As a result, hydrocarbons are
recycled back to methane. For example, the total reactions rates of ethynyl with diacetylene in
10
X. Gu et al.
Jupiter are on the order of 104-5 molecules cm-2 s-1, several orders of magnitude lower than those on
Titan. Recall that the Ion and Neutral mass Spectrometer (INMS) on board the Cassini spacecraft
conducted the first composition measurements of Titan’s ionosphere (53). These measurements are
crucial for constraining the composition and density of minor neutral molecules formed via proton
transfer processes. Here, signal in the INMS at, for instance, m/z = 75 indicates the existence of a
C6H3+ cation, formed via proton transfer to a hitherto unobserved triacetylene molecule. Therefore,
our laboratory results suggest that the neutral triacetylene molecule is expected to be present in
Titan’s atmosphere.
It should be stressed that our findings to form triacetylene (HCCCCCCH) also have strong
implications to the chemistry of the interstellar medium and of combustion flames. Acetylene
(HCCH), diacetylene (HCCCCH) as well as triacetylene (HCCCCCCH) have been detected in the
proto planetary nebula (PPN) RL 618 ( 54 ), an object on the evolutionary stage between the
asymptotic giant branch (AGB) stage among them the carbon star IRC+10216 as a prototype - and
the planetary nebula phase (55). The transition period from an AGB star to the PN phase remains an
elusive stage in stellar evolution, and molecular tracers like triacetylene can play an important role
in untangling this missing link (56). Compared to the low ultraviolet (UV) photon flux from AGB
stars, circumstellar envelopes of PPN are exposed to higher photon fluxes in the UV region of the
electromagnetic spectrum. This in turn will lead to photodissociation of neutral gaseous molecules
among them acetylene yielding atomic hydrogen plus the ethynyl radical. Based on our studies, the
ethynyl radial can react via a single collision event with ubiquitous diacetylene molecules to yield
the triacetylene molecule. This one-step reaction between two neutral species can effectively replace
hitherto postulated, but unstudied ion-molecule reaction sequences thought to be responsible for the
formation of triacetylene in CRL 618 (57). On the other hand, the low level UV photon flux from
AGB stars prohibit a rich photochemistry and hence the synthesis of polyynes such as diacetylene
and triacetylene. Therefore, our studies suggest that the triacetylene molecule can also be utilized as
a tracer to pin down the existence or absence of UV fields strong enough to induce a rich photochemistry in PPNs. Since triacetylene is a key molecule observed in extragalactic sources such as
toward the Large Magellanic Cloud 11 (58), triacetylene can be also employed to trace UV fields in
extragalactic astronomy to identify organic chemistry factories in space (56). This molecule is also
considered to be linked to the formation of polycyclic aromatic hydrocarbons (PAHs) during the
post AGB phase (54,55).
11
X. Gu et al.
To summarize, our investigations provided a unique view into the chemical dynamics on the
reaction of ethynyl radicals with diacetylene. This presents the very first step toward a systematic
and detailed understanding of important elementary reactions involving organic transient species
like diacetylene leading to the formation and complexation of the organic aerosol layer in Titan’s
atmosphere and in hydrocarbon-rich atmospheres of planets like Uranus and Neptune in general
(59). We have identified triacetylene as the sole reaction product under single collision conditions.
This presents the very first experimental detection of the triacetylene reaction product under single
collision conditions as present in Titan’s atmosphere ever. The potential energy surface exhibits no
entrance barrier, the reaction to synthesize triacetylene is exothermic, and all transition states which
are involved are below the energy of the reactant molecules. Therefore, this elementary reaction of
two neutral reactants represents an efficient pathway to produce triacetylene molecules in
hydrocarbon-rich atmospheres of planets and their satellites such as Titan in those regions where
density profiles of photolytically generated ethynyl radicals and diacetylene molecules overlap. Our
findings also hold significant implications for the understanding and quantification of the
atmospheric triacetylene and the aerosol balance. We have demonstrated the key role of the
triacetylene molecule to yield more complex polyynes. Ethynyl radicals (CCH) or 1,4-butadiynyl
radicals (HCCCC), formed via photodissociation of acetylene and diacetylene, respectively (60), can
react with triacetylene to tetraacetylene (C8H2) and even pentaacetylene (C10H2) thus offering crucial
triacetylene sinks and vital polyyne precursors for the organic aerosol layers. Consequently, our
study also foresees the formation and presence of previously unidentified polyynes in Titan’s
atmosphere. We hope that this combined experimental, theoretical, and modeling study will act as a
template and trigger much needed, successive investigation of the central polyyne chemistry such as
of triacetylene and tetraacetylene and more complex systems under realistic, Titan-like simulation
conditions so that a complete picture of the processes involved in the chemical processing of Titan’s
atmosphere will emerge.
12
X. Gu et al.
Materials and Methods
Experimental. The crossed beam experiments are carried out under single collision conditions
utilizing a crossed molecular beams machine described in Reference (61). We generated a pulsed
supersonic ethynyl radical beam, C2D(X2+), in situ via laser ablation of graphite at 266 nm (62)
and seeding the ablated species in neat deuterium which acts as a carrier as well as a reactant gas.
This beam contains also atomic carbon (C(3P)), dicarbon (C2(X1g+;a3), and (C3(X1g+)). However, these species do not hamper with the reactive scattering signal of the reaction of D1-ethynyl
with diacetylene at mass-to-charge ratios of m/z = 75 and 74. Here, the C(3P) + C4H2 (X1g+) reaction leads to reactive scattering signal at m/z = 61 (C5H+), whereas reactive collisions of C2(X1g+;
a3 ) with diacetylene result in the formation of C6H (73 amu) plus atomic hydrogen. The reactions
of tricarbon with closed shell molecules like triacetylene have significant entrance barriers much
higher than the collision energy utilized in the present experiments (63); also, signal from this
reaction is expected to arise at C7H (75 amu), but was not observed experimentally. Consequently,
none of the co-reagents interferes with the reaction of D1-ethynyl radicals plus diacetylene. Note
that molecular deuterium was preferred over molecular hydrogen since molecular hydrogen would
result in the formation of ethynyl (C2H; m/z = 25). Since m/z = 25 can also arise from 13C12C in the
beam, the use of D1-ethynyl (m/z = 26) made it easier to optimize the production conditions of the
D1-ehtynyl radical without interferences from
13
C12C in the beam. The generated D1-ethinyl beam
passes a skimmer; a four-slot chopper wheel located after the ablation zone and prior to the
interaction region selected a segment of the pulse with a peak velocity vp of 2082  50 ms-1 and
speed ratio S of 2.3  0.3. This beam crosses an argon-seeded diacetylene beam (at seeding fractions
of 5 %) in the main chamber at a collision energy of 40.1 kJ mol-1 (vp = 599  20 m s-1, S = 21.0 
1.0). Diacetylene (> 99.5 % purity) was synthesized according to literature (64) and purified by trapto-trap destillation. The reactively scattered species were monitored utilizing a triply differentially
pumped detector consisting of a Brink-type electron-impact-ionizer, quadrupole mass-filter, and a
Daly ion detector by recording time-of-flight spectra (TOF) at different laboratory angles at mass to
charge ratios.
Computational. Our electronic structure calculations were conducted at a level of theory high
enough to predict relative energies of all local minima, transition states, and products of the D1ethynyl reaction with diacetylene to a precision of about 5 kJmol-1. Stationary points were optimized
13
X. Gu et al.
at the hybrid density functional B3LYP level (65,66) with the 6-311G** basis set. Vibrational
frequencies were calculated using the same B3LYP/6-311G** method and were used to compute
zero-point vibrational energy (ZPE) corrections and reaction rate constants without scaling. In the
calculations of vibrational frequencies we used the deuterium isotope for the relevant hydrogen
atom. Relative energies of various species were refined employing the coupled cluster CCSD(T)
method (67) with extrapolation to the complete basis set (CBS) limit and including ZPEs obtained
by B3LYP calculations for all C6H2D intermediates. The results of ab initio calculations, such as
relative energies and molecular parameters, were utilized in RRKM calculations of energydependent rate constants for individual unimolecular steps. The detailed procedure for such
calculations has been described in detail in our previous works (68,50). We computed rate constants
as functions of available internal energy of each intermediate or transition state, where the internal
energy was taken as a sum of the chemical activation energy and the collision energy, assuming that
a dominant fraction of the latter is converted to internal vibrational energy. Rate constants for
decomposition of reaction intermediates were used to evaluate their lifetimes.
Modeling. We perform a one-dimensional diurnally-averaged simulation of the chemical processes
in the atmosphere of Titan. Two complete sets of hydrocarbon chemistry are used. The former set
contains 41 hydrogen and carbon species and about 250 chemical reactions. The latter includes four
additional hydrocarbons and ~200 additional reactions. The concentrations of chemical species are
calculated to steady state by solving the mass continuity equation:
ni  i

 Pi  Li ,
t
z
where ni is the number density for species i, i the vertical flux, Pi the chemical production rate, and
Li the chemical loss rate, all evaluated at time t and altitude z. Pi and Li are calculated based on the
chemical schemes published in the literature (4, 25-27). The vertical flux is given by
n
D
K
T (1   i ) Di  K zz
 i   i ( Di  K zz )  ni ( i  zz )  ni
[
] + wini,
z
H i H atm
z
T
where Di is the species’ molecular diffusion, Hi the species’ scale height, Hatm the atmospheric scale
height, i the thermal diffusion parameter, and T the temperature.
14
X. Gu et al.
ACKNOWLEDGEMENTS. The experimental work was supported by the Chemistry division of
the US National Science Foundation (NSF-CRC CHE-0627854). MCL was supported in part by
NSC grant 97-2628-M-001-001 to Academia Sinica; YLY was supported by NASA grant
NNG06GF33G to Caltech. We thank Dr. Eric Wilson (JPL) for valuable discussions.
15
X. Gu et al.
Figure 1. Selected time-of-flight (TOF) spectra at mass to charge ratio m/z = 75 (C6HD+). Circles
denote experimental data, the solid lines the calculated distributions utilizing the best fit center-ofmass translational and angular distributions (Figure 3).
16
X. Gu et al.
12000
Integrated counts
10000
8000
6000
4000
C.M.
2000
0
0
15
30
45
Lab angle,
60
75
90
(degree)
Figure 2. Laboratory angular distribution of the reactive scattering signal observed at m/z = 75
(C6HD+). The filled circles present the experimental data together with the error limits, the solid
lines the calculated distribution utilizing the best fit center-of-mass translational and angular
distributions (Figure 3). Black balls depict carbon, green balls deuterium, and blue balls hydrogen
atoms. The laboratory angular distribution (LAB) is obtained by integrating the TOF spectra at each
laboratory angle and correcting for distinct data accumulation times.
17
1.2
1.2
1.0
1.0
0.8
0.8
T ()
P (ET)
X. Gu et al.
0.6
0.6
0.4
0.4
0.2
0.2
0.0
0.0
0
20
40
60
80
100
120
140
160
0
180
30
60
90
120
150
180
Center of mass angle,  (degree)
-1
Product translational energy, ET (kJmol )
Figure 3. Center-of-mass translational energy distribution (upper left), center-of-mass angular
distribution (upper right), and center-of-mass velocity flux contour map for the reaction of D1ethynyl (C2D(X2+)) with diacetylene (C4H2(X1g+)) to form atomic hydrogen plus D1-triacetylene
(DCCCCCCH(X1+)) (bottom). The shaded areas of the center-of-mass translational energy and
angular distributions delimit the range of acceptable fits within the error limits, whereas the solid
lines define the ‘best fit’ functions. The contour lines connect points of constant fluxes (yellow:
minimum intensity; red: maximum intensity; the direction of the D1-ethynyl beam is defined as 00
and of the diacetylene beam as 1800).
18
X. Gu et al.
relative energy, kJmol-1
-100
-104
(116)
-123
(-122)
-148
(147)
- 125 (124)
[2]; Cs; 2A'
-159 (-158)
-200
[3]; Cs; 2A'
- 226 (-225)
-300
[1]; Cs; 2A'; - 294 (293)
Figure 4. Calculated potential energy surface of the D1-ethynyl (C2D(X2+)) plus diacetylene
(C4H2(X1g+)) reaction to form D1-triacetylene (DCCCCCCH(X1+)) plus atomic hydrogen. The
energies in parenthesis are the energies for the corresponding ethynyl (C2H(X2+)) plus diacetylene
(C4H2(X1g+)) reaction taken from Reference 49; point groups and electronic wave functions of the
intermediates are also indicated. Black balls depict carbon, dark green balls deuterium, and blue
balls hydrogen atoms.
19
X. Gu et al.
Figure 5. Profiles of ethynyl radical (C2H), acetylene (C2H2), diacetylene (C4H2), triacetylene
(C6H2), and tetraacetylene (C8H2) calculated by the reference model (solid curves), which uses
Moses et al.’s (69) hydrocarbon chemistry and assumes no physical adsorption. Our model that
considers physicochemical adsorption is shown by the dashed curves for comparison.
20
X. Gu et al.
References and Notes
1
Lorenz RD (2006) The Exploration of Titan. Johns Hopkins APL Technical Digest 27:133-142.
2
Lebreton JP, Witasse O, Sollazzo C, Blancquaert T, Couzin P, Schipper AM, Jones JB, Matson
DL, Gurvits LI, Atkinson DH, et al. (2005) Nature 438, 758.
3
Jiang X, Camp CD, Shia R, Noone D, Walker C, Yung YL (2004) Quasi-biennial oscillation
and quasi-biennial oscillation-annual beat in the tropical total column ozone: A two-dimensional
model simulation. J Geophys Res 109:D16305.
4
Roe HG, Bouchez AH, Trujillo CA, Schaller EL, Brown ME (2005) Discovery of Temperate
Latitude Clouds on Titan. Astrophys J Lett 618:L49.
5
Roe H G, de Pater I, McKay CP (2004) Seasonal variation of Titan’s stratospheric ethylene (C2H4)
observed. Icarus 169:440.
6
Griffith CA, Owen T, Miller GA, Geballe T (1998) Transient clouds in Titan’s lower atmosphere.
Nature 395: 575.
7
Flasar FM et al. (2005) Titan's atmospheric temperatures, winds, and composition. Science 308: 975
8
Israel G, Szopa C, Raulin F, Cabane M, Niemann HB, Atreya SK, Bauer SJ, Brun J-F, Chassefiere
E,Coll P, Conde E, Coscia D, Hauchecorne A, Millian P, Nguyen M-J, Owen T, Riedler W,
Samuelson RE, Siguier J-M, Steller M, Sternberg R, Vidal-Madjar C (2005) Complex organic
matter in Titan's atmospheric aerosols from in situ pyrolysis and analysis. Nature 438:796.
9
Sarker N, Somogyi A, Lunine JI, Smith MA (2003) Titan aerosol analogues: Analysis of the
nonvolatile tholins. Astrobiology 3:719.
10
Kerr RA (2005) Planetary Science: Titan, once a world apart, becomes familiar. Science 307:330.
11
Yung YL, Allen M, Pinto JP (1984) Photochemistry of the Atmosphere of Titan – Comparison
between Model and Observations. Astrophys J Suppl S 55:465.
12
Hidayat T, Marten A, Bezard B, Gautier D, Owen T, Matthews HE, Paubert G (1998). Millimeter
and Submillimeter Heterodyne Observations of Titan - The Vertical Profile of Carbon Monoxide
in Its Stratosphere. Icarus 133:109
13
Letourneur B, Coustenis A (1993) Titan’s atmospheric structure from Voyager 2 infrared spectra.
Planet Space Sci 41:593.
21
X. Gu et al.
14
Hidayat T, Marten A, Bezard B, Owen T, Matthews HE, Paubert G (1997) Millimeter and
submillimeter heterodyne observations of Titan: retrieval of the vertical profile of HCN and the
12
15
C/13C ratio. Icarus 126:170.
Lunine JI, Yung YL, Lorenz RD (1999) On the Volatile Inventory of Titan from Isotopic
Abundances in Nitrogen and Methane. Planet Space Sci 47:1291.
16
Gurwell MA (2004) Submillimeter observations of Titan: Global measures of stratospheric
temperature, CO, HCN, HC3N, and the isotopic ratios C-12/C-13 and N-14/N-15. Astrophys J
Lett 616:L7.
17
Wong AS, Morgan CG, Yung YL, Owen T (2002) Evolution of CO on Titan. Icarus 155:382.
18
Rodriguez S, Paillou P, Dobrijevic M, Ruffié G, Coll P, Bernard JM, Encrenaz P (2003) Impact
of aerosols present in Titan’s atmosphere on the Cassini radar experiment. Icarus 164:213.
19
Owen TC (2000) On the origin of Titan’s atmosphere. Planet. Space Sci. 48:747.
20
Coustenis A, Taylor F (1999) Titan – The Earth-Like Moon. (World Scientific, Singapore).
21
Lavvas PP, Coustenis A, Vardavas IM (2008) Coupling photochemistry with haze formation in
Titan's atmosphere, Part I: Model description. Planet Space Sci 56:27.
22
Lara LM, Lellouch E, Lopez-Moreno JJ, Rodrigo R (1996) Vertical distribution of Titan’s
atmospheric neutral constituents. J Geophys Res 101:12151.
23
Gazeau MC, Cottin H, Vuitton V, Smith N, Raulin F (2000) Experimental and theoretical
photochemistry: application to the cometary environment and Titan's atmosphere. Planet Space
Sci 48:437.
24
Keller CN, Anicich VG, Cravens TE (1998) Model of Titan’s ionosphere with detailed
hydrocarbon ion chemistry. Planet Space Sci 46:1157.
25
Banaszkiewicz M, Lara LM, Rodrigo R, Lopez-Moreno JJ, Molina-Cuberos GJ (2000) A coupled
model of Titan’s atmosphere and ionosphere. Icarus 147:386.
26
Lara LM, Lellouch E, Shematovich V (1999) Titan’s atmospheric haze: the case for HCN
incorporation. Astron & Astrophys 341:312.
27
Wilson EH, Atreya SK (2004) Current state of modeling the photochemistry of Titan’s mutually
dependent atmosphere and ionosphere. J Geophys Res 109:E06002.
28
Lorenz RD, McKay CP, Lunine JI (1997) Photochemically driven collapse of Titan’s atmosphere.
Science 275:642.
22
X. Gu et al.
29
Wilson EH, Atreya SK, Coustenis A (2003) Mechanisms for the formation of benzene in the
atmosphere of Titan. J Geophys Res 108:5014.
30
Roush TL, Dalton JB (2004) Reflectance spectra of hydrated Titan tholins at cryogenic
temperatures and implications for compositional interpretation of red objects in the outer Solar
System. Icarus 168:158.
31
Lorenz R, Mitton J (2002) Lifting Titan’s Veil. (Cambridge University Press, Cambridge).
32
Lebonnois S, Bakes E L O, McKay C P (2008) Transition from gaseous compounds to aerosols in
Titan’s atmosphere. Icarus 159:505.
33
Gladstone GR, Allen M, & Yung YL (1996) Icarus 119, 1.
34
Vuitton V, Doussin JF, Benilan Y, Raulin F, Gazeau MC (2006) Experimental and theoretical
study of hydrocarbon photochemistry applied to Titan stratosphere. Icarus 185:287.
35
Clarke DW, Ferris JP (1997) Titan Haze: Structure and Properties of cyanoacetylene and
cyanoacetylene-acetylene photopolymers. Icarus 127:158.
36
Aflalaye A, Sternberg R, Raulin F, Vidal-Madjar C (1995) Gas chromatography of Titan’s
atmosphere VI. Analysis of low-molecular-mass hydrocarbons and nitriles with BPX5 capillary
columns. J Chromatogr A 708:283.
37
Sternberg R, Szopa C, Coscia D, Zubrzycki S, Raulin F, Vidal-Madjar C, Niemann H, Israel G
(1999) Gas chromatography in space exploration capillary and micropacked columns for in situ
analysis of Titan’s atmosphere. J Chromatogr A 846:307.
38
Aflalaye A, Sternberg R, Coscia D, Raulin F, Vidal-Madjar C (1996) Gas chromatography of
Titan’s atmosphere VIII. Analysis of permanent gases with carbon molecular sieve packed
capillary columns. J Chromatogr A 761:195.
39
Rannou P, Hourdin F, McKay CP, Luz D (2004) A coupled dynamics-microphysics model of
Titan’s atmosphere. Icarus 170:443
40
Ramirez SI, Navarro-Gonzales R, Coll P, Rauli F (2001) Possible contribution of different energy
sources to the production of organics in Titan's atmosphere. Adv Space Res 27:261.
41
Waite JH Jr., Young DT, Cravens TE, Coates A J, Crary FJ, Magee B, Westlake J (2007) The
Process of Tholin Formation in Titan 's Upper Atmosphere. Science 316:870.
42
Liang MC, Yung YL, Shemansky DE (2007) Photolytically generated aerosols in the mesosphere
and thermosphere of Titan. Astrophys J Lett 661:L199.
23
X. Gu et al.
43
Weis MS (1986) Ph.D. thesis. (University of California, Berkeley).
44
Levine RD, Bernstein RB (1987) Molecular Reaction Dynamics and Chemical Reactivity.
(Oxford University Press, Oxford).
45
Miller WB, Safron SA, Herschbach DR (1967) Exchange reactions of alkali atoms with alkali
halides. Collision complex mechanism. Discuss Faraday Soc 44:108.
46
Sattelmeyer KW, Stanton JF (2000) Computational Studies of C6H2 Isomers. J Am Chem Soc
122:8220.
47
It should be stressed that although our collision energies are higher than the relevant conditions in
Titan’s atmosphere, the formation of triacetylene is, as confirmed by our calculations, the only
accessible channel, even down to very low temperatures of 90 K – 200 K as prevailing in Titan’s
atmosphere.
48
Liang MC, Shia RL, Lee AYT, Allen M, Friedson AJ, & Yung YL (2005) Astrophysical
Journal 635, L177-L180
49
Landera A, Krishtal S P, Kislov VV, Mebel A M, Kaiser RI (2008) Theoretical Study of the C6H3
Potential Energy Surface and Rate Constants and Product Branching Ratios of the C2H + C4H2
and C4H+ C2H2 Reactions. J Chem Phys 128:214301.
50
Smith IWM, Herbst E, Chang Q (2004) Rapid neutral-neutral reactions at low temperatures: a
new network and first results for TMC-1. MNRAS 350:323.
51
Zhou L, Kaiser RI, Gao LG, Chang AHH, Liang MC, & Yung YL (2008) Astrophysical Journal
686, 1493.
52
Moses JI, Fouchet T, Bezard B, Gladstone GR, Lellouch E, & Feuchtgruber H (2005) Journal Of
Geophysical Research-Planets 110.
53
Vuitton, V., Yelle, R.V., McEwan, M.J. (2007) Ion chemistry and N-containing molecules
in Titan’s upper atmosphere. Icarus, 191: 722.
54
Cernicharo J. Heras AM, Tielens AGGM, Pardo JR, Herpin F, Guelin M, Waters LBFM (2001)
Infrared space observatory's discovery of C4H2, C6H2, and benzene in CRL 618. Astrophys J Lett
546:L123.
55
Ziurys LM (2006) The chemistry in circumstellar envelopes of evolved stars: following the origin
of the elements to the origin of life. PNAS 103:12274.
56
Sun K (2006) Synthesis of organic compounds in planetary nebulae and proto-planetary nebulae.
24
X. Gu et al.
AIP Conference Proc. 855:154
57
Woods PM, Millar TJ, Herbst E, Zijlstra AA (2003) The chemistry of protoplanetary nebulae.
Astron Astrophys 402:189.
58
Bernard-Salas J, Peeters E, Sloan GC, Cami J, Guiles S, Houck JR (2006) The Spitzer IRS
spectrum of SMP LMC 11. Astrophys J Lett 652:L29.
59
Recently, diacetylene was also observed in Neptune’s atmosphere at the 0.1 mbar level at mixing
ratios of 3 ± 1 × 10-12. Meadows, V.S., Orton, G., Line, M., Liang, M.C., Yung, Y.L., an Cleve, J.,
Nurgdorf, M.J. (2008). First Spitzer observations of Neptune: Detection of new hydrocarbons.
Icarus 197: 585.
60
Silva R, Gichuhi WK, Huang C, Doyle MB, Kislov VV, Mebel AM, Suits AG (2008) H
elimination and metastable lifetimes in the UV photoexcitation of diacetylene. PNAS 105:12713.
61
Gu X, Guo Y, Zhang F, Mebel AM, Kaiser RI (2006) Reaction Dynamics of Carbon-Bearing
Radicals in Circumstellar Envelopes of Carbon Stars. Faraday Discuss 133:245.
62
Gu X, Guo Y, Kawamura E, Kaiser RI (2006) Characteristics and Diagnostics of a Ultrahigh
Vacuum Compatible Laser Ablation Source for Crossed Molecular Beam Experiments. J Vac Sci
Technol A, 24:505.
63
Guo Y, Gu X, Zhang F, Mebel AM, Kaiser RI (2007) A Crossed Molecular Beam Study on the
Formation of Hexenediynyl Radicals (H2CCCCCCH; C6H3 (X2A')) via Reactions of Tricarbon
Molecules, C3(X1Σg+), with Allene (H2CCCH2; X1A1) and Methylacetylene (CH3CCH; X1A1).
Phys Chem Chem Phys 9:1972. Gu X, Guo Y, Mebel AM, Kaiser RI (2007) A Crossed Beams
Investigation of the Reactions of Tricarbon Molecules, C3, with Acetylene, C2H2, Ethylene, C2H4,
and Benzene C6H6. Chem Phys Lett 449:44.
64
Jones AV (1952) Infrared and Raman spectra of biacetylene. Proc R Soc London Ser A 211:285
Okabe H (1981) Photochemistry of acetylene at 1470. Ang Chem Phys 75:2772.
65
Becke AD (1993) Density-functional thermochemistry. III. The role of exact exchange. J Chem
Phys 98:5648.
66
Lee C, Yang W, Parr RG (1988) Development of the Colle-Salvetti correlation-energy formula
into a functional of the electron density. Phys Rev B 37:785.
67
Purvis GD, Bartlett RJ (1982) A full coupled-cluster singles and doubles model: The inclusion of
disconnected triples. J Chem Phys 76:1910. Scuseria GE, Janssen CL, Schaefer HF  (1988) An
25
X. Gu et al.
efficient reformulation of the closed-shell coupled cluster single and double excitation (CCSD)
equations. J Chem Phys 89:7382. Scuseria GE, Schaefer HF  (1989) Is coupled cluster singles
and doubles (CCSD) more computationally intensive than quadratic configuration interaction
(QCISD)? J Chem Phys 90:3700. Pople JA, Head-Gordon M, Raghavachari K (1987) Quadratic
configuration interaction. A general technique for determining electron correlation energies. J
Chem Phys 87:5968.
68
Kislov VV, Nguyen TL, Mebel AM, Lin SH, Smith SC (2004) Photodissociation of benzene
under collision-free conditions: An ab initio/Rice–Ramsperger–Kassel–Marcus study. J Chem
Phys 120:7008.
69
Moses JI, Bezard B, Lellouch E, Gladstone GR, Feuchtgruber H, & Allen M (2000) Icarus
143: 244.
26