Acrylonitrile characterization and high energetic photochemistry at Titan temperatures

Icarus xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Icarus journal homepage: www.elsevier.com/locate/icarus

Acrylonitrile characterization and high energetic photochemistry at Titan temperatures A. Toumi, N. Piétri ⇑, T. Chiavassa, I. Couturier-Tamburelli ⇑ Aix-Marseille Université, CNRS, PIIM, UMR 7345, 13397 Marseille, France

a r t i c l e

i n f o

Article history: Received 10 July 2014 Revised 3 October 2014 Accepted 22 October 2014 Available online xxxx Keywords: Titan Photochemistry Ices, IR Spectroscopy

a b s t r a c t Laboratory infrared spectra of amorphous and crystalline acrylonitrile (C2H3CN) ices were recorded between 4000 and 650 cm1. Heating up the acrylonitrile sample to 160 K shows details on the transition between amorphous and crystalline ice at 94 K. This molecule can be used as an indicator of the surface temperature of Titan since it is known also to be 94 K. The desorption energy of acrylonitrile was determined using two methods (IRTF and mass spectrometries) to be around 35 kJ mol1. Solid phase acrylonitrile was irradiated with vacuum ultraviolet (VUV) light at low temperatures (20, 70, 95 and 130 K) using a microwave-discharge hydrogen flow lamp. Isoacrylonitrile, cyanoacetylene (HC3N), isocyanoacetylene (HC2NC), acetylene (C2H2) and hydrogen cyanide (HCN) were identified as photoproducts by using FTIR spectroscopy. The branching ratio of each pathway has been calculated for the different temperatures. We have estimated the acrylonitrile, HCN and HC3N mC„N stretching band strengths to 1 be respectively A ¼ 3:98  1018 , A ¼ 1:38  1018 and A ¼ 2:92  1018 cm molecule . Ó 2014 Elsevier Inc. All rights reserved.

1. Introduction During these last 30 years, the chemical atmospheric composition of Titan (the largest satellite of Saturn) has been extensively explored (Liang et al., 2007; Kammer et al., 2013). The different constituents have been identified by infrared spectroscopy during the Voyager and Cassini-Huygens missions (Coustenis et al., 1989, 1991, 2007). A wide variety of organic molecules comes from the coupled photochemistry of CH4 and N2, the two most abundant components in the atmosphere. Among the molecules detected, numerous hydrocarbon species have been found in gas phase: acetylene (C2H2), ethylene (C2H4), ethane (C2H6), methyl-acetylene (CH3CCH), propane (C3H8), diacetylene (C4H2) and benzene (C6H6) (Coustenis et al., 2007; Hébrard et al., 2007). Other interesting compounds formed in the atmosphere are the nitriles as hydrogen cyanide (HCN), cyanoacetylene (HC3N), cyanogen (C2N2), dicyanoacetylene (C4N2) and more recently detected, acrylonitrile (C2H3CN) (Cui et al., 2009; Magee et al., 2009; Lellouch et al., 2010). Some of them (HC3N, C4N2, and CH3 CH2CN) have been observed in solid phase such as C2H2 (Khanna, 2005a,b). As stated by Moore et al. (2010), these interpretations of Titan observations depend on the knowledge of the spectra of various molecular solids ⇑ Corresponding authors. E-mail addresses: [email protected] (N. Piétri), isabelle.couturier@ univ-amu.fr (I. Couturier-Tamburelli).

suspected to be present. The temperature and physical state (different phases) of the medium are important for the molecular identification since they can induce large frequency shifts in the absorption band position, intensity and width of the absorption (Khanna, 2005a). Spectra have been published concerning crystalline acrylonitrile (or 2-propenenitrile or vinyl cyanide) (Dello Russo and Khanna, 1996), but no data have been found concerning the different phases and complete photochemistry of acrylonitrile. Encouraged by the works of Khanna et al. (Khanna, 2005a,b; Dello Russo and Khanna, 1996; Samuelson et al., 1997) and Kim and Kaiser (2009) who suggested that nitriles could be condensable on Titan, it seems to be interesting to study the acrylonitrile molecule. Since Titan’s surface is known to be cold 94 K and lower temperatures around 70 K are present in the low atmosphere or troposphere (50 km), acrylonitrile could also exist in its solid form as either an amorphous or crystalline solid because its freezing point is at 190 K in normal conditions (Finke et al., 1972). Since the dominant energy source available to dissociate N2 and CH4 in the upper atmosphere is solar radiation (Lavvas et al., 2011), the neutral species present in Titan’s atmosphere can undergo dissociation. So after providing measurements of amorphous and crystalline condensed solid acrylonitrile at different temperatures in this work, we will present the results of photolysis experiments on acrylonitrile. These experiments have been conducted in order to understand the behavior of the molecules when they are formed and submitted to UV radiation in Titan’s atmosphere. The

http://dx.doi.org/10.1016/j.icarus.2014.10.042 0019-1035/Ó 2014 Elsevier Inc. All rights reserved.

Please cite this article in press as: Toumi, A., et al. Acrylonitrile characterization and high energetic photochemistry at Titan temperatures. Icarus (2014), http://dx.doi.org/10.1016/j.icarus.2014.10.042

2

A. Toumi et al. / Icarus xxx (2014) xxx–xxx

quantification of gas-phase acrylonitrile photochemistry has already been studied at 213.9 nm by Gandini and Hackett (1978). Two primary processes have been identified in the gas phase with the main one corresponding to the CAH and CAC bonds cleavage yielding to C2H2 and HCN formation. The branching ratio of this process was estimated to be 50%. The pathway leading to HC3N (with the loss of molecular hydrogen) yields 31%. Wilhelm et al. (2009), using time-resolved Fourier transform infrared emission spectroscopy, studied the photodissociation of acrylonitrile and its isotopologue (CD2@CDCN) at 193 nm. The presence of HCN and its isomer HNC with C2H2 has been revealed during these experiments. Recently we reported (Toumi et al., 2014) that the photochemistry of acrylonitrile in argon matrix induces the formation of C2H2:HCN, C2H2:HNC complexes, HC3N (with three other isomers) and isoacrylonitrile. The purpose of this work is to present the first experimental approach for the photochemistry of acrylonitrile in simulated atmospheric conditions of Titan (CouturierTamburelli et al., 2014). Herein we report temperature-dependent infrared spectra of amorphous and crystalline ices of acrylonitrile. We present the results obtained after temperature-dependent photolysis of solid acrylonitrile. Using FTIR, mass spectrometry and previous experiments, we characterized the different products obtained in solid phase and compared the results with similar processes measured in the gas phase. We give a rough approximation of the branching ratios obtained for each product identified during photolysis experiment at different temperatures. Since all the identified compounds have a mC„N or mN@C stretching band, we used these modes to determine the branching ratio. We also discuss the consequence of this study on the understanding of acrylonitrile photochemical behavior in titanian conditions. 2. Experimental section Acrylonitrile (from Aldrich, with a purity P99%) was used after purification by vacuum distillation. Acrylonitrile was vapor deposited at a rate of 6  101 mol min1 on a gold-plated copper surface kept at different temperatures between 20 and 300 K with the help of a cold head cryostat (CTI, model 21) within a high vacuum chamber (ca 107 mbar). The warming up of the samples was performed at a heating rate between 0.8 and 5 K min1 using a resistive heater along with a Lakeshore model 331 temperature controller. The spectra were recorded in reflection–absorption (doubled absorption) using a Fourier Transform Infrared Spectrometer (Nicolet serie II Magna System 750) from 4000 to 650 cm1. Each spectrum was averaged over one hundred scans with 1 cm1 resolution. Mass spectra of the samples were recorded up to 130 amu with a resolution of 1 amu during precisely controlled warm-up sessions using a quadrupole mass spectrometer (MKS Microvision-IP plus), with a 70 eV electron impact ionization source. A microwave-discharge hydrogen flow lamp (Opthos Instruments, operating pressure 0.4 mbar, MgF2 window) was used as the farUV source; its flux, confined to the range of 3–10 eV, is dominated by two broad bands, centered around 120 (Lyman-a) and 160 nm, with a continuum in the visible. We estimated the Lyman-a photons flux using the actinometric method (Gerakines et al., 2000) in which we photolyzed a pure O2 solid deposition with the hydrogen flow lamp and measured the photolysis rate when O2 turned into O3. The Lyman-a photons flux was calculated to be 4.02 (±1.15)  1012 photons cm2 s1 with a forward power of 70 W and a reflected part less than 2 W. Quantitatively, a difference is observed with the solar flux Lyman-a photons penetrating the high atmosphere of Titan and which is known to be 4.12  1011 photons cm2 s1 (Toublanc et al., 1995; Lean, 1991). A recent work performed by Chen et al. (2014) permits us to calculate the total flux of the lamp from the Lyman-a photons one. With a F-Type microwave McCarroll cavity and with a MgF2 window, the

proportion of Lyman-a is 8.4%. So, in our case, we have a total flux of 4.79 (±1.37)  1013 photons cm2 s1. Irradiation of solid acrylonitrile was also carried out using an Osram 200 W high-pressure mercury lamp equipped with a quartz envelope (k > 230 nm). The photon flux of this last lamp is estimated to be 2.75  1016 photons cm2 s1 (Gudipati et al., 2013; Couturier-Tamburelli et al., 2014). Since the reactants are consumed during the photoreaction, the band integration strengths are used to monitor the increase of the products infrared band intensities and to estimate how much of each product is formed. The amount of initial acrylonitrile molecules was obtained with the mC„N stretching mode at 2229 cm1. Since the band strength of mC„N of pure acrylonitrile is unknown, it was necessary to determine this value. To do so, we deposited different mixtures of acrylonitrile and CO2 (Gerakines, 1995). The absorption strength of the mC„N stretching mode of pure acrylonitrile at 20 K was calculated from the data experiment to be 1 A ¼ 3:98  1018 cm molecule using the areas under the C„N stretching mode peak and the absorption of the mCO mode of CO2 (Gerakines, 1995). This value is of the same order of magnitude as the one published by Bernstein et al. for acrylonitrile in H2O 1 which is A ¼ 7:7  1018 cm molecule (Bernstein et al., 1997). We used the same procedure to determine the band strengths of HCN and HC3N which are listed in Table 1. Based on the work of Hudson and Moore, 2004, the band strength used for CN stretching of nitriles and isonitriles are assumed to be equivalent. For example, the value used in our experiments for the isoacrylonitrile is the one that we have calculated for the acrylonitrile molecule (3.98  1018 cm molecule1). R ~Þdm ~ in cm1 could be An infrared band’s integrated intensity sðm converted to a molecular column density N, in molecule cm2, R ~ ~ through sðmAÞdm provided that the band’s intrinsic strength, A in cm molecule1, was known. 2.1. Infrared spectra of solid acrylonitrile 2.1.1. Amorphous ices formed at 20 K Vibrational spectroscopy of acrylonitrile in rare-gas matrices has been studied earlier (Toumi et al., 2014). Here we present full infrared spectrum of pure solid acrylonitrile at different temperatures (20, 95, 130 K) along with the vibrational assignment by comparing with earlier studies (Dello Russo and Khanna, 1996; Khlifi et al., 1999). These comparisons are summarized in Table 2. In our experiment, the spectrum of solid amorphous acrylonitrile is obtained at 20 K in the 4000–650 cm1 range and shows fundamental and combination modes (Table 2). The most intense bands are observed at 970 and 2229 cm1 and have been assigned, respectively to wagging modes noted m12, m13 and mC„N stretching mode noted m4. In all the spectral regions, the absorptions of amorphous ice are broad and clumpy. 2.1.2. Annealing and phase transition to crystalline ice The pure acrylonitrile ice is heated from 20 to 180 K with different heating rates (from 0.8 to 2 K min1 when followed by IR and

Table 1 Band strengths of CN stretching mode for different nitriles related to our experiments. Band strength (1018 cm molecule1) A A A A A

(acrylonitrile, pure) (acrylonitrile, water) (Bernstein et al., 1997) (HC3N, pure) (HCN, pure) (HCN, water) (Bernstein et al., 1997)

3.98 7.7 2.92 1.38 5.1

Please cite this article in press as: Toumi, A., et al. Acrylonitrile characterization and high energetic photochemistry at Titan temperatures. Icarus (2014), http://dx.doi.org/10.1016/j.icarus.2014.10.042

3

A. Toumi et al. / Icarus xxx (2014) xxx–xxx Table 2 Experimental frequencies and relative intensities of solid acrylonitrile at different temperatures and in argon matrix (Toumi et al., 2014). Assignments

m1 CH stretching m2 CH stretching m3 CH stretching Combination band

m6 + m9 m4 CN stretching m12 + m13 m12 + m14/m13 + m14 m5 C@C stretching m6 CH deformation m7 CH rocking m8 CH2 rocking m12 H2C@C wagging m13 C@CHACN wagging m9 CAC stretching m14 C@C torsion

Wavenumber (cm1) Argon matrix (Toumi et al., 2014) 15 K (rel. intensity)

Amorphous ice

Crystalline ice

Solid 20 K (rel. intensity) (our work)

Solid 95 K (rel. intensity) (our work)

Solid 130 K (rel. intensity) (our work)

Solid 35 K after 130 K (Dello Russo and Khanna, 1996)

3124 (3) 3074 (14) 3042 (14) 2998 (–) 2277 (3) 2235 (15) 1907 (17) 1651 (<1) 1616 (7) 1413 (54) 1287 (<1) 1097 (2) 974 (54) 953 (100) 867 (3) 683 (7)

3114 (5) 3069 (9) 3031 (4) 2990 (3) 2281 (1) 2229 (30) 1957 (9) 1668 (2) 1608 (2) 1414 (14) 1285 (2) 1092 (7) 970 (100)

3111 (43) 3075 (22) 3033 (7) 2991 (19) 2294 (7) 2233 (82) 1962 1667 (9) 1610 (4) 1425 (39) 1294/1283 (3) 1106 (6) 963 (100)

3114 (56) 3069 (21) 3030 (17) 2991 (18) 2293 (6) 2228 (74) – 1670 (7) 1607 (6) 1421 (30) 1298 (1) 1100 (4) 985/977 (100)

3113/3101 (47) 3068/3051 (21) 3028 (14) – – 2228 (65)

871 (2) 689 (13)

874 (2) 691 (10)

880 (3) 693 (9)

from 0.5 to 3.5 K min1 by mass). During annealing, we observed two steps in the crystallization process, a first one at 95 K and a second at 130 K. In Dello Russo and Khanna’s work (Dello Russo and Khanna, 1996), CH2@CHCN was deposited at 15 K. Then, the solid phase was annealed at 130 K in order to obtain a crystalline sample and then cooled back to 35 K before recording the spectrum. All these absorption bands are summarized in Table 2. We can note that the crystalline phase obtained by Dello Russo and Khanna is similar with the solid phase obtained in our experiment at 130 K. Upon crystallization, the spectrum of the solid phase is transformed into structured and well defined bands. The spectra of the two most intense absorption bands of acrylonitrile at 20, 95 and 130 K are presented in Fig. 1. Between 90 and 95 K, crystallization of acrylonitrile results in a rapid change in position and shape of the different bands: m4 shifts to a higher frequency (2233 cm1) while m13 is shifted to a lower frequency (963 cm1). It is a coincidence that this temperature is close to the one of Titan’s surface (about 94 K). The annealing above 95 K promotes further changes. A second step in the crystallization process occurs around 130 K. The m4 mode shifts to 2233 cm1 by 95 K then returns to 2228 cm1 by 130 K and increases in intensity (Fig. 1). The m13 is shifted to higher frequency after the second step of crystallization. 2.1.3. Sublimation of acrylonitrile The desorption process can be followed as depicted in Fig. 2 showing the m4 integrated band intensity variations. Between 95 and 125 K, the peak intensity is unchanged, which is indicative of a stable adsorption state. At 135 K, we observe an abrupt decrease in intensity resulting from the sublimation of solid acrylonitrile. The desorption’s temperature has been measured around 145 K (Fig. 2). The acrylonitrile desorption follows a zeroth-order kinetic model (Ostwald, 1887). In order to determine the desorption energy of acrylonitrile, we performed a Thermal Programmed Desorption (TPD) study where approximately 1  1018 molecules were deposited for each experiment. The evolution of the number of molecules for a zeroth-order kinetic is given by:



dNðtÞ ¼ kðTÞ dt

where N(t) is the number of molecules and k(T) is the temperaturedependant kinetic rate of the considered reaction (here desorption).

– 1606/1585 (7) 1428/1422/1419/1413 (31) 1299 (1) 1104/1100 (10) 993/987 982/979/975 (100) 881/875 (8) 694 (18)

By applying Arrhenius’ law (Arrhenius, 1889) and integrating with time, we have:

  Edes t NðtÞ ¼ N0  m0  exp  RT If we take into account the fact that the temperature follows a linear increase (T = T0 + bt), the last relation becomes:

    Edes T  T0  NðTÞ ¼ N0  m0  exp  RT b T0 and b are the initial temperature and heating rate respectively. m0 is the zeroth-order pre-exponential term, Edes is the desorption energy and R is the ideal gas constant. The first part of the desorption curves obtained from the IR signals is fitted to the following expression:

HðTÞ ¼

  NðTÞ 1 T  T0 Edes   exp  ¼ 1  m0  NðT ¼ T 0 Þ No b RT

where H(T) is the surface coverage (relative integrated absorbance of the m4 band) of the solid phase molecules at the temperature T. For normalization, the first order pre-exponential term (1013 s1) is multiplied by a 1015 molecules cm2 and by a 1 cm2 factors to respect the dimensions in the surface coverage formula and to give the zeroth-order pre-exponential term, m0. If we fix this value to 1028 molecules cm2 s1, we obtain Edes = 34.86 (±0.79) kJ mol1 by averaging the different results of each TPD experiment and the ±0.79 kJ mol1 uncertainty is the result of the dispersion of Edes from different TPD experiments. This Edes value is lower than the one evaluated with TPD for HC3N 39 (±8) kJ mol1 (Borget et al., 2001) and C4N2, 42 (±5) kJ mol1 (Guennoun, 2004; Guennoun et al., 2005a) adsorbed on amorphous ice, and could be included in astrochemistry codes leading to the residence time (tres) of a molecule on a grain or on the aerosols. The solid acrylonitrile desorption has also been followed with a quadrupole mass spectrometer for different heating rates like illustrated in Fig. 2. These experiments have been performed in order to confirm the results obtained from the IR study. In the case of a mass spectrum and for a zeroth-order kinetic desorption, we have to fit the rising edge of the desorption rate K curve with a zeroth-order of the Polanyi–Wigner equation (Readhead, 1962; Carter, 1962) in which the desorption rate K is a function of the temperature T:

Please cite this article in press as: Toumi, A., et al. Acrylonitrile characterization and high energetic photochemistry at Titan temperatures. Icarus (2014), http://dx.doi.org/10.1016/j.icarus.2014.10.042

A. Toumi et al. / Icarus xxx (2014) xxx–xxx

2260

T=130K T=95K T=20K

ν4

(c)

(b) (a)

Absorbance

3,00E+016

0,4

0,00E+000

130

140

0,0 150

Temperature (Kelvin) 2250

2240

2230

2220

2210

2200

T=130K T=95K T=20K

ν12, ν13

(c)

(b) (a)

1040 1020 1000 980

960

940

920

900

880

860

840

Wavenumber (cm-1) Fig. 1. Infrared spectra of acrylonitrile in the 2260–2200 and 1040–840 regions: (a) at 20 K (amorphous phase), (b) at 95 K (first crystalline phase) and (c) at 130 K (second crystalline phase).

KðTÞ ¼ 

0,8

120

Wavenumber (cm-1) 0,26 0,24 0,22 0,20 0,18 0,16 0,14 0,12 0,10 0,08 0,06 0,04 0,02 0,00 -0,02

1,2 6,00E+016

Surface coverage

0,60 0,55 0,50 0,45 0,40 0,35 0,30 0,25 0,20 0,15 0,10 0,05 0,00 -0,05

Desorption rate (molecules cm-2)

Absorbance

4

  dN m0 Edes ¼  exp  dT b RT

We decided to let only Edes as a free parameter and to fix the m0 factor to 1028 s1. In the case of desorption mass spectra, we obtain Edes = 35.23 (±0.42) kJ mol1. As for the IR study, the uncertainty for Edes is a result of the dispersion from the different TPD experiments. The value obtained with mass spectrometry is in good agreement with the result of the IR study. The difference may be caused by the difference of sensibility between the two methods. For the rest of this paper, we will consider an average value of 35.05 (±0.61) kJ mol1 as the desorption energy for acrylonitrile. 2.2. Photodegradation of pure acrylonitrile As reported earlier (Toumi et al., 2014), the photochemistry of acrylonitrile trapped in argon matrix induces the formation of C2H2 trapped with HCN or HNC, HC3N and its isomers and the formation of isoacrylonitrile. In Titan’s environment, acrylonitrile could be found in solid phase in the aerosols of the atmosphere where they are exposed to UV radiation. The photochemistry of acrylonitrile is performed at different temperatures and different wavelengths in order to simulate its photoreactivity at different altitudes. 2.2.1. Infrared spectroscopy for products attributions and quantifications The acrylonitrile ice was first irradiated at k > 230 nm with a mercury lamp for several hours at 70 K in order to simulate the

Fig. 2. Integrated normalized absorbance of acrylonitrile m4 mode with temperature (b = 1.4 K min1) determined by FTIR spectrometry (cross points) and evolution of the desorption rate (b = 3.5 K min1) of acrylonitrile with temperature determined by mass spectrometry (circle points) for samples deposited at 80 K.

photochemistry of the lower atmosphere (troposphere). According to the UV spectrum published by Eden et al. (2003), no detectable change in the infrared spectra was observed by the photolysis at this wavelength. With its characteristic transitions [r ? r⁄ (172.5 nm), p ? p⁄ (203 nm) and n ? p⁄ bands (217 nm)] and also its Rydberg transitions in the 115–146 nm region, it seemed interesting to photolyze the acrylonitrile at k > 120 nm with a microwave-discharge hydrogen flow lamp. We performed this experiment at different temperatures: 20 K, 70 K (when the acrylonitrile is still in its amorphous phase and at the lower temperature measured in the atmosphere) and after the two transition phases (at 95 and 130 K). At 70 K, we observed the decrease of the acrylonitrile absorption bands and the growth of numerous bands (Fig. 3). As shown in Fig. 4, photolyzing with a hydrogen lamp during 1020 min results in 75% of acrylonitrile depletion. Comparing to previous experiments (Guennoun et al., 2005a), the bands which growth at 3217, 2266 and 2064 cm1 are attributed to the solid HC3N. Its isomer, the HC2NC (2204 and 2032 cm1) has also been put in evidence during this photodegradation process. We note the formation of two bands located at 2133 and 2086 cm1 that grow until the end of photolysis. On the basis of the work developed by Hudson and Moore (Hudson and Moore, 2004) and our previous work (Toumi et al., 2014), the band at 2133 cm1 can be attributed to isoacrylonitrile in solid phase. For this product, two other bands are also observed at 1616 and 1109 cm1. The band observed at 2086 cm1 is attributed to the formation of HCN by comparison with previous experiments of nitriles in solid phase (Guennoun et al., 2005b, 2006). HCN formation implies the formation of the C2H2 molecule. The band observed at 3214 cm1 could be attributed to mCAH of HCN and C2H2. The experiments have been performed at different temperatures and the same products are obtained but not in the same proportions than at 20 K and 95 K. At 130 K, no clear product absorption bands were detected: only bands attributed to refractory compounds were observed.

2.2.1.1. Warming of photolyzed acrylonitrile ice. After UV irradiation of acrylonitrile films at different temperatures, the samples were warmed up to 300 K with a 2 K min1 rate during which infrared spectra were measured continuously. We observed a decrease in intensity of all the photoproducts bands which disappear when acrylonitrile desorption occurs around 145 K. At room temperature, only small infrared bands associated with a residue remain (Fig. 5). Some of these infrared features are characteristic of mCAH, mC„N and mC@C (2940, 2242 and 1639 cm1). This residue, which

Please cite this article in press as: Toumi, A., et al. Acrylonitrile characterization and high energetic photochemistry at Titan temperatures. Icarus (2014), http://dx.doi.org/10.1016/j.icarus.2014.10.042

5

A. Toumi et al. / Icarus xxx (2014) xxx–xxx

0,030

isoacrylonitrile

HC 3 N

HCN

-CN -NC

1020 minutes 420 minutes deposition

HC3N HC2NC

0,025

0,0025

(c)

Absorbance

Absorbance

0,0020 (b)

0,020 0,015 acrylonitrile

0,010

-C=C -C=N Aromac CH2 CH3

-CH

0,0015

0,0010

0,0005

0,005 (a)

0,000

0,0000

2200

2100

2000

1900

1800

2800

-1

2600

2400

2200

2000

1800

1600

1400

Wavenumber (cm-1)

Wavenumber (cm ) Fig. 3. FTIR spectra at 70 K: (a) of pure acrylonitrile, (b) after 420 min of VUV irradiation and (c) after 1020 min of VUV irradiation.

Fig. 5. Infrared spectra of non-volatile residue obtained at room temperature produced by k > 120 nm photochemistry of solid acrylonitrile at 70 K.

is white-yellow, is not similar in color to those obtained during the photolysis experiment of another nitrile (C4N2) detected in the Titan’s atmosphere (Couturier-Tamburelli et al., 2014; Gudipati et al., 2013). In this last experiment, the residue observed is orange like the Titan’s haze. Nevertheless, in the two experiments the residue can be dissolved in methanol.

non-irradiated acrylonitrile. In this case, the formation of HC3N from a non-irradiated sample is due to the loss of two hydrogen atoms caused by the ionization. We have also to take into account the fact that acrylonitrile and HC3N will sublime at the same temperature. In order to make sure that HC3N has also been produced by the photolysis experiment (and not only by ionization of acrylonitrile sample), we have to compare the relative intensities of m/ z = 51 to m/z = 53 calculated from both experiments (sublimation of irradiated and non-irradiated acrylonitrile). Because of the difference between these ratios, we are able to confirm that, in the photolysis experiment, a part of the m/z = 51 peak comes from photochemically-produced HC3N. With the same procedure, we are able to confirm the formation of HCN (m/z = 27) and C2H2 (m/ z = 26) due to photolysis and not only as the result of the electronic impact ionization of acrylonitrile. The ratio between the desorption experiments with or without irradiation must be taken at a different temperature for HCN and C2H2 because of their lower sublimation temperature compared to the acrylonitrile and HC3N ones.

2.2.1.2. Mass spectrometry experiment. In order to confirm the products formed during this experiment, we performed a complementary analysis using mass spectrometry. Since the m/z of isoacrylonitrile is the same than for acrylonitrile (m/z = 53), its formation can’t be confirmed by mass spectrometry. The mass spectrum is monitored between 20 and 300 K at different heating rates. During the temperature ramp, several masses are recorded using quadrupole mass spectrometer, as shown for m/z = 53 in Fig. 2 with a 3.5 K min1 heating rate. The mass spectrum is monitored at 147 K when acrylonitrile sublimates (Fig. 6). The spectrum profiles obtained after irradiation were also compared with the spectrum of the same ions resulting from warming non-irradiated acrylonitrile. For the irradiated sample, we are able to identify HC3N (m/z = 51). Due to the electron impact ionization induced by this method (mass spectrometry), it is clear that we can observe a peak located at m/z = 51 after the sublimation of

Fitting equation: N(t)= N0 × e

N acrylonitrile (molecules cm-2)

3,50E+016

-k×t

k= 3.66 (±0.24) × 10-3 min-1

t1=2 ¼

3,00E+016

R²= 0.99652 2,50E+016

2,00E+016

1,50E+016

1,00E+016

5,00E+015

0

200

400

600

800

1000

2.2.2. Photochemical considerations 2.2.2.1. Photodissociation cross section. After 1020 min of irradiation at 70 K, approximately 75% of the acrylonitrile has been consumed as shown in Fig. 4. The evolution of the absorbance band located at 2229 cm1 as a function of the time is fitted by a first order kinetic rate for the photodissociation. The half life t1/2 (Ostwald, 1887) given by

1200

Irradiation time (minutes) Fig. 4. Evolution of the column density of acrylonitrile during VUV irradiation at 70 K. The column density was calculated from the m4 stretching mode and the decrease is accounted by an exponential decay.

lnð2Þ k

is determined graphically around 292 min (or 17,520 s). The photodissociation rate (k) is written as k = r  f, where r represents the photodissociation cross-section for a given species and f the ultraviolet radiation flux of our hydrogen lamp (f = 4.79 (±1.37)  1013 photons cm2 s1). The kinetic rate of acrylonitrile’s photodissociation reaction (k) is found to be 3.96 (±0.18)  105 s1. So, the photodissociation cross-section obtained in this present work for a photolysis at 70 K is calculated to be racrylonitrile = 8.27 (±2.73)  1019 cm2 photon1 and the other values for different temperatures are listed in Table 3. We also calculated the photo-dissociation cross-section at 20 K which is not a Titan-relevant temperature and estimated it to be 6.62 (±1.97)  1019 cm2 photon1. 2.2.2.2. Determination of the quantum yields. Based on the knowledge of the band strength we are able to determine the number

Please cite this article in press as: Toumi, A., et al. Acrylonitrile characterization and high energetic photochemistry at Titan temperatures. Icarus (2014), http://dx.doi.org/10.1016/j.icarus.2014.10.042

6

A. Toumi et al. / Icarus xxx (2014) xxx–xxx

Relative Intensity (a.u.)

1,0

0,8

0,6

0,4

0,2

0,0 20

30

40

50

60

m/z Fig. 6. Mass spectrum recorded at 147 K after VUV irradiation of solid acrylonitrile performed at 70 K (heating rate is 5 K min1).

Table 3 Photodissociation cross-sections calculated for 70, 95 and 130 K. Temperature (K)

racrylonitrile (cm2 photon1)

70 95 130

8.27 (±2.73)  1019 1.66 (±5.32)  1018 3.01 (±1.15)  1017

Five major compounds were identified. The quantum yields of three products (C2H2/HCN, isoacrylonitrile and HC3N) have been measured from their respective infrared bands after 1020 min irradiation time and Scheme 1 displays the pathways leading to 3 major compounds. The quantum yield of isocyanoacetylene HC2NC has not been calculated because the quantity formed is not sufficient to quantify properly the amount. Among photochemistry processes found at k > 120 nm in solid phase, the main one corresponds to a CAC bond cleavage followed by an isomerization process. The branching ratio of this process at 70 K, which induces the isoacrylonitrile formation, is estimated to be 24%. The pathway producing HCN yields around 16%. Finally, the loss of H2 producing HC3N represents around 10%. The quantum yield deficit could be explained by polymer formation which is the source of a non-volatile residue (Fig. 5). Experiments have been carried out at 20, 70 and 95 K. We have completed these experiments to compare the effects of temperature on photolyzed acrylonitrile with the same irradiation times. When the irradiations are performed at different temperatures, the same photoproducts are obtained and confirm the photodissociation pathway. Only the branching ratios are different, like illustrated in Table 4. All the experiments induce the formation of the same main product, corresponding to the isomerization of acrylonitrile. At 20 K, the branching ratio leading to isoacrylonitrile is estimated at 60%, this latter falls down to 20% at 95 K. We can see a drastic decrease of these photoproducts with the increase of temperature. So the non-observation of these compounds during the 130 K experiment is not surprising. 3. Discussion

of molecules for specific absorption bands. The ratio between the products molecules formed and the acrylonitrile molecules consumed is the quantum yield /:

/¼

Nðproduct formedÞ Nðacrylonitrile consumedÞ

Using acrylonitrile, one of the nitriles detected in Titan’s atmosphere as a model, we conducted experiments in order to understand the photochemistry of Titan’s condensed aerosols at upper altitudes where the high energy photons arrive. Most of the investigations of Titan’s constituents have focused on the gas phase.

Acetylene (C2H2)

Hydrogen cyanide (HCN)

Cyanoacetylene (HC3N)

~ 10%

~ 16%

~ 24%

Isoacrylonitrile CH2=CHNC Scheme 1. The main products obtained during the photolysis of acrylonitrile at k > 120 nm. The branching ratios are presented for the 70 K experiment.

Please cite this article in press as: Toumi, A., et al. Acrylonitrile characterization and high energetic photochemistry at Titan temperatures. Icarus (2014), http://dx.doi.org/10.1016/j.icarus.2014.10.042

7

A. Toumi et al. / Icarus xxx (2014) xxx–xxx Table 4 Branching ratios estimated at different temperatures during acrylonitrile photolysis experiments. Photoproducts

Isoacrylonitrile (%) HC3N (%) HCN (%)

Temperature (K) 20

70

95

60 17 22

24 10 16

20 1 7

Some compounds have been found in solid phase on Titan in the polar regions. C4N2 was detected by Khanna and Samuelson et al. (Khanna et al., 1987; Samuelson et al., 1997) via its m8 absorption band at 478 cm1. The comparison between laboratory and Titan data lead to suggest also the presence of HCN (Mayo and Samuelson, 2005) and HC3N condensates. Here we suggest that acrylonitrile may be present in solid phase on Titan. Like mentioned by Kim and Kaiser (2009), since the phase transition of acrylonitrile is observed around 94 K, we can use the phase transition as an indicator of the surface temperature. As mentioned by numerous authors, nitriles are one of the better choices among the candidates for detecting condensed species. Actually, three of them (HC3N, C4N2 and CH2CH3CN) have already been observed in solid phase (Khanna et al., 1987; Khanna, 2005a,b). Thus, acrylonitrile is likely to be condensed in the atmosphere at T < 140 K. So, if acrylonitrile is present at T < 140 K in solid phase in the atmosphere, its abundance depends on the behavior when it is submitted to radiation. As stated by Lavvas et al. (2008), the main pathway of photodissociation lead to the formation of HC3N in gas phase. Our results suggest that the main photodissociation pathway of solid acrylonitrile is CAC cleavage estimated to be 24% at 70 K. This branching ratio was not determined in gaseous phase, because of the non-obtention of isoacrylonitrile in these experimental conditions (Gandini and Hackett, 1978; Wilhelm et al., 2009). Our results show that when the CAC bond is cleaved, it promotes the elimination process to form isoacrylonitrile. For the two other pathways which are observed in gas phase leading to HCN and HC3N formation, the branching ratios are lower in the solid phase. In this last state, the isomerization is the most favorable process. In our experiment, the radicals are trapped in the solid phase allowing the recombination and isomerization processes to occur, like in argon matrix (Toumi et al., 2014). The time necessary to destroy acrylonitrile (t1/2) must be compared with the residence time (tres) of acrylonitrile on the aerosols. The previously calculated desorption energy leads us to estimate the residence time (tres) of a molecule on an aerosol (Sandford and Allamandola, 1993)

t res ¼

1

m1 expð ERTdes Þ

where m1 is the surface molecule oscillator frequency. Here we used the standard value for m1 which is 3  1012 s1 (Sandford and Allamandola, 1993). Residence times of acrylonitrile are given in Table 5 and it varies between 1.51  106 years at 70 K and 0.54 s at 150 K. In laboratory conditions, the photodissociation half-life time of acrylonitrile is found to be 292 min at 70 K and 8 min at 130 K. For this last temperature, we may have a contribution from the desorption process of acrylonitrile. The photodissociation halflife times are also listed in Table 5. So, at 70 K solid acrylonitrile can be destroyed by photolysis within one day whereas it is destroyed in less than a few minutes at 130 K. For altitude above 100 km, in the high atmosphere, the temperature is higher than 150 K and acrylonitrile will desorb in less than a second if it is trapped on an aerosol. This molecule could be submitted to short UV (k > 110 nm) and long UV (k > 300 nm)

Table 5 Residence time (tres) and photodissociation half-life time (t1/2) of acrylonitrile determined for different characteristic temperatures of Titan atmosphere. Temperature (K)

tres

t1/2 (min)

70 95 130 150

1.51  106 years 72 days 40 s 0.54 s

292 145 8 –

irradiations when it is trapped on the aerosols in the upper and in the lower Titan atmosphere respectively. The solar flux Lyman-a photons arriving in the upper atmosphere is estimated to be 4.12  1011 photons cm2 s1 (Toublanc et al., 1995; Lean, 1991). Since the Lyman-a flux represents 70% of the total solar spectrum, we can estimate the total solar flux to be 5.89  1011 photons cm2 s1. With this last flux and since our lamp has a flux of 4.79  1011 photons cm2 s1, we can estimate the photodissociation half-life time for acrylonitrile at 130 K to be 8 min. So this molecule should be observable in the gas phase in the upper atmosphere where it will be then photodissociated. The residence time values are obtained in laboratory conditions, under a high vacuum (107 mbar), but it is important to keep in mind that the pressure in Titan’s atmosphere is known to be 1.5 bar at the surface (Coustenis et al., 1991) and will be lower than 103 mbar in the upper atmosphere. Consequently, in titanian conditions, the residence times are inevitably longer. To summarize, if the acrylonitrile is present in the upper atmosphere with short UV photons and temperatures above 150 K, the total desorption occurs in less than a second and then the molecule couldn’t undergo the photochemistry in solid phase, on aerosols, even taking into consideration longer residence times due to the higher atmospheric pressure. Thus, in the upper atmosphere, the products formed by photochemistry of acrylonitrile ice (HC3N, isoacrylonitrile, HCN) can’t be observed. So, no part of HC3N and HCN detected in the high atmosphere of Titan can come from the photochemistry of solid acrylonitrile. On the other hand, when the acrylonitrile is present in the lower part of the atmosphere, it may be subjected to long UV irradiations (k > 300 nm) but this molecule does not dissociate in this range of wavelengths. At these temperatures, included between 70 K (for the coolest) and 95 K (temperature of the surface), the acrylonitrile desorbs between 72 days and 106 years approximately. So, when the acrylonitrile is formed in gas phase in the upper part of the atmosphere, it could condense in the lower portions of atmosphere and could exist in solid form. Thus acrylonitrile could be used like an indicator of the surface temperature (Kim et al., 2010; Kim and Kaiser, 2009). Based on our study we recommend utilizing infrared spectrum of solid acrylonitrile to determine if acrylonitrile exists in the crystalline or amorphous form in the aerosol particles in Titan’s atmosphere. 4. Conclusion Several experiments concerning solid phase acrylonitrile have been performed in order to analyze the behavior of this molecule when it is present in the atmosphere of Titan. In this work, we measured the desorption energy, the residence time, and also the photodissociation cross-section of acrylonitrile at different temperatures. The values are useful for atmospheric chemistry models of Titan as well as for stars-forming region models. Different photodissociation pathways have been determined and branching ratio measurements have shown that the pathway leading to isoacrylonitrile was dominant. Taking into account the residence time of acrylonitrile in solid form in the lower and upper atmosphere of Titan and irradiation conditions of this molecule in the solid phase, we have shown that

Please cite this article in press as: Toumi, A., et al. Acrylonitrile characterization and high energetic photochemistry at Titan temperatures. Icarus (2014), http://dx.doi.org/10.1016/j.icarus.2014.10.042

8

A. Toumi et al. / Icarus xxx (2014) xxx–xxx

it can be protected from solar radiation (k < 300 nm) in the lower atmosphere and then be used as a probe to determine the temperature of the surface due to the phase transition (T = 94 K) which produces very characteristic changes in the infrared spectrum. We also showed that this molecule could be photodissociated in solid phase by VUV radiation that play an important role in the upper atmosphere of Titan. However, due to the high temperatures prevailing in the upper atmosphere (T > 150 K), the residence time of acrylonitrile in the solid state is very short and it will be irradiated in the gas phase by VUV radiation. So, any photoreactivity of acrylonitrile would occur from its gas phase in the atmosphere of Titan. Acknowledgments This work has been funded by the French national interdisciplinary program ‘‘Environnements Planétaires et Origines de la Vie’’ (EPOV). We also wish to acknowledge Dr Fabrice Duvernay for his precious help about experimental details. References Arrhenius, S.A., 1889. On the reaction velocity of the inversion of cane sugar by acids. Z. Phys. Chem. 4, 226–248. Bernstein, M.P., Sandford, S.A., Allamandola, L.J., 1997. The infrared spectra of nitriles and related compounds frozen in Ar and H2O. Astrophys. J. 476, 932– 942. Borget, F., Chiavassa, T., Allouche, A., Marinelli, F., Aycard, J.P., 2001. Cyanoacetylene adsorption on amorphous and crystalline water ice films: Investigation through a matrix and a quantum study. J. Am. Chem. Soc. 123, 10668–10675. Carter, G., 1962. Thermal resolution of desorption energy spectra. Vacuum 12, 245– 254. Chen, Y.-J. et al., 2014. Vacuum ultraviolet emission spectrum measurement of a microwave-discharge hydrogen-flow lamp in several configurations: Application to photodesorption on CO Ice. Astrophys. J. 781, 15–28. Coustenis, A., Bézard, B., Gautier, D., 1989. Titan’s atmosphere from Voyager infrared observations: I. The gas composition of Titan’s equatorial region. Icarus 80, 54–76. Coustenis, A., Bézard, B., Gautier, D., Marten, A., Samuelson, R., 1991. Titan’s atmosphere from Voyager infrared observations: III. The vertical distributions of hydrocarbons and nitriles near Titan’s North Pole. Icarus 89, 152–167. Coustenis, A., Achterberg, R., Conrath, B., Jennings, D., Marten, A., Gautier, D., Bjoraker, G., Nixon, C., Romani, P., Carlson, R., et al., 2007. The composition of Titan’s stratosphere from Cassini/CIRS mid-infrared spectra. Icarus 189, 35–62. Couturier-Tamburelli, I., Gudipati, M.S., Lignell, A., Jacovi, R., Piétri, N., 2014. Spectroscopic studies of non volatile residue formed by photochemistry of solid C4N2: A model of condensed aerosol formation on Titan. Icarus 234, 81–90. Cui, J., Yelle, R.V., Vuitton, V., Waite Jr., J.H., Kasprzak, W.T., Gell, D.A., Niemann, H.B., Müller-Wodarg, I.C.F., Borggren, F., Fletcher, G.G., et al., 2009. Analysis of Titan’s neutral upper atmosphere from Cassini Ion Neutral Mass Spectrometer measurements. Icarus 200, 581–615. Dello Russo, N., Khanna, R.K., 1996. Laboratory infrared spectroscopic studies of crystalline nitriles with relevance to outer planetary systems. Icarus 123, 366– 395. Eden, S., Limao-Vieira, P., Kendall, P., Mason, N.J., Hoffmann, S.V., Spyrou, S.M., 2003. High resolution photo-absorption studies of acrylonitrile, C2H3CN and acetonitrile, CH3CN. Eur. Phys. J. 26, 201–210. Finke, H.L., Messerly, J.F., Todd, S.S., 1972. Thermodynamics properties of acrylonitrile, 1-aminopropane, 2-aminopropane, and 2-methyl-2aminopropane. J. Chem. Thermodyn. 4, 359–374. Gandini, A., Hackett, P.A., 1978. The photochemistry of acrylonitrile vapour at 213.9 nm. Can. J. Chem. 56, 2096–2098. Gerakines, P.A., 1995. The infrared band strengths of H2O, CO and CO2 in laboratory simulations of astrophysical ice mixtures. Astron. Astrophys. 296, 810–818. Gerakines, P.A., Moore, M.H., Hudson, R.L., 2000. Carbonic acid production in H2O:CO2 ices UV photolysis vs. proton bombardment. Astron. Astrophys. 357, 793–800.

Gudipati, M.S., Jacovi, R., Couturier-Tamburelli, I., Lignell, A., Allen, M., 2013. Photochemical activity of Titan’s low-altitude condensed haze. Nat. Commun. 4, 1648. http://dx.doi.org/10.1038/ncomms2649. Guennoun, Z., 2004. Thesis of Université de Provence. Guennoun, Z., Couturier-Tamburelli, I., Piétri, N., Aycard, J.P., 2005a. Interstellar ice surface site modification induced by dicyanoacetylene adsorption. J. Phys. Chem. B 109, 3437–3441. Guennoun, Z., Piétri, N., Couturier-Tamburelli, I., Aycard, J.P., 2005b. Photoreactivity of cyanoacetylene trapped in water ice: An infrared, isotopic and theoretical study. J. Phys. Chem. A 109, 8299–8305. Guennoun, Z., Piétri, N., Couturier-Tamburelli, T., Aycard, J.P., 2006. UV photolysis of C4N2 in water ices: New possible route of synthesis of ammonium bicarbonate and ammonium formate. J. Phys. Chem. A 110, 7738–7743. Hébrard, E., Dobrijevic, M., Bénilan, Y., Raulin, F., 2007. Photochemical kinetics uncertainties in modeling Titan’s atmosphere: First consequences. Planet. Space Sci. 55, 1470–1489. Hudson, R.L., Moore, M.H., 2004. Reactions of nitriles in ices relevant to Titan, comets, and the interstellar medium: Formation of cyanate ion, ketenimines, and isonitriles. Icarus 172, 466–478. Kammer, J.A., Shemansky, D.E., Zhang, X., Yung, Y.L., 2013. Composition of Titan’s upper atmosphere from Cassini UVIS EUV stellar occultations. Planet. Space Sci. 88, 86–92. Khanna, R.K., 2005a. Condensed species in Titan’s atmosphere: Identification of crystalline propionitrile (C2H5CN, CH3CH2CN) based on laboratory infrared data. Icarus 177, 116–121. Khanna, R.K., 2005b. Condensed species in Titan’s stratosphere: Confirmation of crystalline cyanoacetylene (HC3N) and evidence for crystalline acetylene (C2H2) on Titan. Icarus 178, 165–170. Khanna, R.K., Perera-Jarmer, M.A., Ospina, M.J., 1987. Vibrational infrared and Raman spectra for dicyanoacetylene. Spectrochim. Acta 43A, 421–425. Khlifi, M. et al., 1999. Absolute intensities of the infrared bands of gaseous acrylonitrile. J. Mol. Spectrosc. 194, 206–210. Kim, Y.S., Kaiser, R.I., 2009. An Infrared spectroscopic study of amorphous and crystalline ices of vinylacetylene and implications for Saturn’s Satellite Titan. Astrophys. J. Suppl. Ser. 181, 543–547. Kim, Y.S., Bennett, C.J., Chen, L.H., O’Brien, K., Kaiser, R.I., 2010. Laboratory studies of the irradiation of solid ethane analog ices and implications to Titan’s chemistry. Astrophys. J. 711, 744–756. Lavvas, P.P., Coustenis, A., Vardavas, I.M., 2008. Coupling photochemistry with haze formation in Titan’s atmosphere, Part I: Model description. Planet. Space Sci. 56, 27–66. Lavvas, P., Galand, M., Yelle, R.V., Heays, A.N., Lewis, B.R., Lewis, G.R., Coates, A.J., 2011. Energy deposition and primary chemical products in Titan’s upper atmosphere. Icarus 213, 233–251. Lean, J., 1991. Variations in the Sun’s radiative output. Rev. Geophys. 29, 505–535. Lellouch, E. et al., 2010. Sounding of Titan’s atmosphere at submillimeter wavelengths from an orbiting spacecraft. Planet. Space Sci. 58, 1724–1739. Liang, M.C., Yung, Y.L., Shemansky, D.E., 2007. Photolytically generated aerosols in the mesosphere and thermosphere of Titan. Astrophys. J. 661, L199–L202. Magee, B.A., Waite, J.H., Mandt, K.E., Westlake, J., Bell, J., Gell, D.A., 2009. INMSderived composition of Titan’s upper atmosphere: Analysis methods and model comparison. Planet. Space Sci. 57, 1895–1916. Mayo, L.A., Samuelson, R.E., 2005. Condensate clouds in Titan’s North Polar stratosphere. Icarus 176, 316–330. Moore, M.H., Ferrante, R.F., Moore, W.J., Hudson, R., 2010. Infrared spectra and optical constant of nitrile ices relevant to Titan’s atmosphere. Astrophys. J. 191, 96–112. Ostwald, F.W., 1887. Lehrbuch der Allgemeinen Chemie. Engelmann. Readhead, P.A., 1962. Thermal desorption of gases. Vacuum 12, 203–211. Samuelson, R.E., Mayo, L.A., Knuckles, M.A., Khanna, R.J., 1997. C4N2 ice in Titan’s North Polar stratosphere. Planet. Space Sci. 45, 941–948. Sandford, S.A., Allamandola, L.J., 1993. H2 in interstellar and extragalactic ices: Infrared characteristics, ultraviolet production, and implications. Astrophys. J. 409, 65–68. Toublanc, D., Parisot, J.P., Brillet, J., Gautier, D., Raulin, F., McKay, C.P., 1995. Photochemical modeling of Titan’s atmosphere. Icarus 113, 2–26. Toumi, A., Couturier-Tamburelli, I., Chiavassa, T., Piétri, N., 2014. Photolysis of astrophysically-relevant acrylonitrile: A matrix experimental study. J. Phys. Chem. A 118, 2453–2462. Wilhelm, M.J., Nikow, M., Letendre, L., Dai, H.L., 2009. Photodissociation of vinyl cyanide at 193 nm: Nascent product distributions of the molecular elimination channels. J. Chem. Phys. 130, 44307–44312.

Please cite this article in press as: Toumi, A., et al. Acrylonitrile characterization and high energetic photochemistry at Titan temperatures. Icarus (2014), http://dx.doi.org/10.1016/j.icarus.2014.10.042