SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental Procedure
  5. Quantum-Chemical Calculation
  6. Results and Discussion
  7. Conclusions
  8. Acknowledgments
  9. Editorial Handling
  10. References

We experimentally studied hydrogen (H)–deuterium (D) substitution reactions of solid methylamine (CH3NH2) under astrophysically relevant conditions. We also calculated the potential energy surface for the H–D substitution reactions of methylamine isotopologues using quantum chemical methods. Despite the relatively large energy barrier of more than 18 kJ mol−1, CH3NH2 reacted with D atoms to yield deuterated methylamines at 10 K, suggesting that the H–D substitution reaction proceeds through quantum tunneling. Deuterated methylamines reacted with H atoms as well. On the basis of present results, we propose that methylamine has potential for D enrichment through atomic surface reactions on interstellar grains at very low temperatures in molecular clouds. D enrichment would occur in particular in the methyl group of methylamine.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental Procedure
  5. Quantum-Chemical Calculation
  6. Results and Discussion
  7. Conclusions
  8. Acknowledgments
  9. Editorial Handling
  10. References

Methylamine (CH3NH2) was recently found in comet-exposed aerogel and foil recovered from comet Wild 2 by Stardust; analyses confirmed it to be inherent to comet Wild 2 (Glavin et al. 2008), as well as the simplest amino acid glycine (Elsila et al. 2009). However, the origin(s) of cometary methylamine remain(s) unknown at present. One of the possible origins is in interstellar molecular clouds, the birth places of stars. Interstellar methylamine was first detected in 1974 toward Sgr B2 and Ori A through ground-based millimeter-wave observations (Fourikis et al. 1974; Kaifu et al. 1974). Several reactions have been proposed for the formation of interstellar methylamine. For example, a series of ion-molecule reactions and dissociative recombination (inline image) is thought to occur in the gas phase (Leung et al. 1984; Herbst 1985; Maeda and Ohno 2006). Photolysis of gas mixtures (CH4, NH3, and H2O) also yields methylamine (Ogura et al. 1988). Methylamine may also be formed on interstellar grains by photolysis of CH4/NH3 ice (Gardner and McNesby 1980) and/or sequential hydrogenation of hydrogen cyanide (HCN) (Woon 2002; Theule et al. 2011). In any case, it is likely that the CH3NH2 molecule is retained on icy grains at as low as 10 K and is subjected to various processes on them.

When methylamine is in/on icy grains, it can be a precursor of more complex molecules upon the surface processes. The surface reactions of methylamine were recently studied in detail in laboratories under astrophysically relevant conditions (Holtom et al. 2005; Bossa et al. 2008, 2009, 2012; Moon et al. 2008; Lee et al. 2009). These laboratory studies are closely related to investigations of the formation of the amino acid glycine (NH2CH2COOH) in interstellar clouds. In cases of both CH3NH2/CO2 mixed-ice irradiation by electrons (Holtom et al. 2005) and CH3NH2/CO2/H2O mixed-ice ultraviolet irradiation (Lee et al. 2009), glycine formation was confirmed through infrared (IR) spectroscopy, and through reactive ion scattering and low-energy sputtering techniques, respectively, although Bossa et al. (2012) claim that the hypothetical yield of glycine is very low. On the other hand, the formation of methylcarbamic acid (CH3NHCOOH), an isomer of glycine, is favored when CH3NH2/CO2 mixed ice is heated to just above 40 K (Bossa et al. 2008).

Deuterium (D) fractionation of interstellar molecules is an important issue when considering chemical evolution in molecular clouds. Some interstellar species were found to be highly enriched in D. Even multiply deuterated species, such as d2-formaldehyde (D2CO) and d3-methanol (CD3OH), have been detected in several interstellar sources (e.g., Ceccarelli et al. 1998; Parise et al. 2004). D enrichment of interstellar species is driven by two processes: gas-phase and grain-surface reactions. Gas-phase chemistry of the D enrichment of interstellar species has been studied extensively (e.g., Roberts and Miller 2000; Gerlich and Schlemmer 2002; Osamura et al. 2005), and the importance of grain-surface reactions for D enrichment has recently become well understood through various experimental studies (Nagaoka et al. 2005, 2007; Hidaka et al. 2009; Oba et al. 2012) and theoretical models (Stantcheva and Herbst 2003; Taquet et al. 2012). Fourikis et al. (1977) reported the probable detection of deuterated methylamine CH3NHD toward Sgr B2; however, it was later not detected in the same target by another group (MacLeod et al. 1979). Although deuterated methylamine such as CH2DNH2 and CH3NHD has never been detected in space, it should be reasonable to consider that interstellar methylamine is deuterated to some extent. Nondetection of deuterated methylamine may be due to its low fractional abundance. A laboratory spectroscopy revealed that CH2DNH2 has a band in the 8–74 GHz region (Tamagake and Tsuboi 1974). Some of other isotopologues such as CH3NHD and CH3ND2 are considered to be observed in the similar frequency region (Takagi and Kojima 1971; Fourikis et al. 1977). Therefore, we expect that extensive survey in this microwave region will result in the finding of deuterated methylamine isotopologues in future astronomical observations. As for chemical reactions toward the D enrichment of interstellar methylamine, no experimental and theoretical studies have been performed so far. Nevertheless, we propose that methylamine could become enriched in D through grain-surface reactions at low temperatures as the case for CH3OH, which has been shown to become D-enriched by surface reactions with D atoms at 10 K (Nagaoka et al. 2007). If this is the case, the D enrichment may be inherited to interstellar glycine and its isomers, which could be produced from methylamine.

In the present study, we focus on D enrichment of methylamine through surface reactions and perform both experimental and theoretical studies of the hydrogen isotopic fractionation of solid methylamine through reactions with D or hydrogen (H) atoms at low temperatures. A possible methylamine D/H ratio after low-temperature surface reactions in molecular clouds will be proposed as an application of the present study.

Experimental Procedure

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental Procedure
  5. Quantum-Chemical Calculation
  6. Results and Discussion
  7. Conclusions
  8. Acknowledgments
  9. Editorial Handling
  10. References

Experiments were performed using the Apparatus for SUrface Reaction in Astrophysics (ASURA) system. ASURA mainly consists of a main chamber and two atomic sources, with multiple turbo-molecular pumps. Details of this apparatus have been described previously (Watanabe et al. 2006; Nagaoka et al. 2007). Four methylamine isotopologues, CH3NH2 (>99% purity; Mitsubishi Gas Chemical Company, Inc.), CH3ND2 (98% purity; Isotec), CD3NH2 (99% purity; Isotec), and CD3ND2 (99% purity; Isotec) were used in the present study. CH3NH2 was further purified to remove impurities by cryogenic distillation, while the other methylamines were used without further purification. Gaseous methylamine was introduced into the main chamber through a capillary plate and deposited onto an aluminum (Al) substrate mounted in the center of the main chamber to produce solid methylamine on its surface. D and H atoms were produced by separate atomic sources through dissociation of D2 and H2, respectively, in a microwave-discharge plasma. Atoms were transferred via a series of poly(tetrafluoroethylene) and a cold Al tube, which was connected to a helium (He) refrigerator. After many collisions with the inner wall of the cold Al pipe, atoms are cooled to the pipe temperature of 100 K (Nagaoka et al. 2007). The fluxes of D and H atoms were measured, using the method reported by Hidaka et al. (2007), to be both approximately 1 × 1014 atoms cm−2 s−1.

The compositions of the solid samples were monitored in situ by reflection–absorption-type Fourier-transform IR (FTIR) spectroscopy with a resolution of 4 cm−1 in the spectral range between 4000 and 800 cm−1. Infrared absorption spectra of the four methylamine isotopologues obtained at 10 K are shown in Fig. 1. The peak assignments for solid CH3NH2, CD3NH2, CH3ND2, and CD3ND2 were derived from Durig et al. (1968) and Durig and Zheng (2001); see Table 1. The desorbed species from the substrate were mass-analyzed with a quadrupole mass spectrometer (QMS). The temperature of the substrate, which was connected to another He refrigerator, was kept at 10 K during exposure for each experiment. After D or H exposure of solid methylamine, temperature-programmed desorption (TPD) spectra were obtained with the QMS at a heating rate of 4 K min−1.

Table 1. Peak positions of solid methylamine isotopologues and integrated band strengths
MoleculePeak position (cm−1)AssignmentbBand strength (cm molecule−1)c
ObservedLiteraturea
  1. a

    Durig et al. (1968).

  2. b

    Durig et al. (1968); Durig and Zheng (2001).

  3. c

    Band strength for CH3NH2 is derived from Holtom et al. (2005).

CH3NH2 33483332NH2 antisymmetric stretch 
  32873260NH2 symmetric stretch 
  31893191H-bonding 
  2969   
  29452942CH3 antisymmetric stretch 
  2923   
  2899   
  28842881CH3 antisymmetric stretch 
  2865   
  27942793CH3 symmetric stretch 
  16181651NH2 deformation 
  14791500CH3 antisymmetric deformation 
  14571467CH3 antisymmetric deformation 
  14211441CH3 symmetric deformation 
  13391353CH3 rock/NH2 twist 
  11611182NH2 twist/CH3 rock1.5 × 10−18
  10451048CN stretch 
  9971005NH2 wag 
  931955NH2 wag 
      
CD3NH2 33463325NH2 antisymmetric stretch 
  3291   
  31883180H-bonding 
  22742277CD3 symmetric deformation 
  22222216CD3 antisymmetric stretch 
  21972194CD3 antisymmetric stretch 
  21472151CD3 antisymmetric deformation 
  21142115CD3 antisymmetric deformation 
  20612058CD3 symmetric stretch 
  1633   
  16051602NH2 deformation3.1 × 10−18
  12871283CD3 rock 
  11411139CD3 symmetric deformation8.7 × 10−19
  10731058CD3 antisymmetric deformation 
  10301014CN stretch 
  953948CD3 rock/NH2 twist 
  830840NH2 wag 
      
CH3ND2 2969   
  29462941CH3 antisymmetric stretch 
  2917   
  2899   
  28832883CH3 symmetric stretch 
  2863   
  28022801CH3 antisymmetric stretch1.6 × 10−18
  25002498ND2 antisymmetric stretch 
  24462443ND2 symmetric stretch 
 ~23642347D-bonding 
  14761468CH3 antisymmetric deformation 
  14561448CH3 antisymmetric deformation 
  14221421CH3 symmetric deformation 
  12211220ND2 deformation2.1 × 10−18
  11321130CH3 rock 
  1030   
  10031005CN stretch 
      
CD3ND2 24992485ND2 antisymmetric stretch 
  24392445ND2 symmetric stretch 
  23632354D-bonding 
  2243   
  22252241CD3 symmetric deformation 
  21952190CD3 antisymmetricstrech 
  21472141CD3 antisymmetric deformation 
  21112105CD3 antisymmetric deformation 
  20662061CD3 symmetric stretch2.4 × 10−18
  20302028CD3 symmetric deformation/CD3 rock 
  12151217ND2 deformation 
  11211121CD3 antisymmetric deformation 
  10751070CD3 antisymmetric deformation 
  10591050CD3 symmetric deformation/CD3 rock 
  941942CN stretch 
  924919CD3 rock/ND2 twist 
image

Figure 1. Infrared spectra of solid CH3NH2, CD3NH2, CH3ND2, and CD3ND2 deposited at 10 K. Peak assignments are derived from Durig et al. (1968), Durig and Zheng (2001), and Bossa et al. (2008). Abbreviations st. and def. indicate stretch and deformation, respectively.

Download figure to PowerPoint

A solid methylamine layer with the thickness of approximately 30 monolayers (ML; 1 ML is equivalent to 1015 molecules cm−2) was exposed to D or H atoms for up to 2 h. A detailed explanation of the quantification of solid methylamine isotopologues is provided below. The integrated band strength of solid CH3NH2 has been reported previously (Holtom et al. 2005; Bossa et al. 2008), while that of solid methylamine isotopologues has yet to be reported. We estimated the integrated band strengths of pure solid CH3ND2, CD3NH2, and CD3ND2 at 10 K using a similar procedure to that for determining the integrated band strengths of pure solid methanol isotopologues such as CH2DOH and CD3OH (Nagaoka et al. 2007). Briefly, pure CH3NH2 was first vapor-deposited under constant gas pressure onto the cold substrate through the capillary plate in the main chamber. The column density of the deposited CH3NH2 was calculated from the peak area and the previously published band strength, as described in Hidaka et al. (2007). The band strength used was 1.5 × 10−18 cm molecules−1 for the NH2 twist/CH3 rock band at 1161 cm−1 (Holtom et al. 2005). Next, we performed the same operation by employing methylamine isotopologues. For CD3NH2, for example, pure CD3NH2 was vapor-deposited under the same conditions (gas pressure, deposition time). The amounts of each isotopologue deposited on the substrate were confirmed to be almost the same based on their TPD spectra. By comparing the peak area of the IR spectrum of CD3NH2 with that of CH3NH2, the approximate integrated band strength of CD3NH2 was determined at 3.1 × 10−18 and 8.7 × 10−18 cm molecules−1 for its NH2 deformation (1605 cm−1) and CD3 symmetric deformation (1141 cm−1), respectively. The approximate integrated band strengths of other methylamine isotopologues determined here are listed in Table 1.

Quantum-Chemical Calculation

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental Procedure
  5. Quantum-Chemical Calculation
  6. Results and Discussion
  7. Conclusions
  8. Acknowledgments
  9. Editorial Handling
  10. References

To clarify the possible processes in the present experiments, we calculated the potential energy surfaces of H abstraction reactions and H–D exchange reactions of methylamine using quantum-chemical methods. We used the hybrid density functional B3LYP method with a 6-311G(d,p) basis set to obtain the molecular structures and zero-point vibrational energies at the energy minima and transition states on the potential energy surfaces. The relative energies were calculated using the CCSD(T) method with the aug-cc-pVTZ basis functions (Dunning 1989). All computations were carried out with the Gaussian 03 program (Frisch et al. 2004).

Results and Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental Procedure
  5. Quantum-Chemical Calculation
  6. Results and Discussion
  7. Conclusions
  8. Acknowledgments
  9. Editorial Handling
  10. References

Exposure of Solid Methylamine (CH3NH2, CD3NH2, or CH3ND2) to D Atoms

Figure 2 shows an infrared absorption spectrum of solid CH3NH2 and the difference spectra obtained after D exposure to CH3NH2 for up to 30 min. Peaks below and above the baseline in the difference spectra represent decreases of CH3NH2 and increases of new species, respectively. Variations in the difference spectra clearly indicate that solid CH3NH2 was consumed concurrently with the formation of some new species after D exposure. By comparing the positions of the new peaks with literature values (Durig et al. 1968), we found that the peaks at 1214 cm−1, 2000–2300 cm−1, and 2300–2500 cm−1 are derived from ND2 deformation, CD3 stretching, and ND2 stretching bands of D-substituted methylamines, respectively. Exposure of CH3NH2 to D2 molecules did not induce any change in the spectrum, indicating that H–D substitution of solid CH3NH2 actually occurred via surface reactions with D atoms at 10 K. Detailed identification of new peaks to D-substituted methylamines such as CH2DNH2 and CHD2NH2 was not possible because of the lack of their standard gases. Figure 3 shows TPD spectra of solid CH3NH2 before and after exposure to D atoms. In the spectrum for intact CH3NH2 before exposure to D atoms (Fig. 3), no signal was observed for mass 36, which indicates an absence of CD3ND2 (mass 36) in the sample. On the other hand, in the spectrum after exposure to D atoms, the presence of CD3ND2 was clearly indicated by the appearance of a mass signal at approximately 110 K (Fig. 3). Therefore, we are convinced that CH3NH2 was fully deuterated after exposure to D atoms.

image

Figure 2. a) Infrared spectrum of solid CH3NH2 before exposure to D atoms, and b) variations in the difference spectrum after exposure to D atoms for up to 30 min. The peaks at approximately 2700 cm−1 were derived from the inherent noise caused by vibration of the He refrigerator.

Download figure to PowerPoint

image

Figure 3. TPD spectra (m/z = 36) of solid CH3NH2 before and after exposure to D atoms.

Download figure to PowerPoint

The following are possible processes for H–D substitution of solid CH3NH2 under the present experimental conditions: a combination of H-abstraction and D-addition reactions (process A) and formation of intermediate species by D addition followed by elimination of H atoms from the methyl and/or amino groups (process B).

Process A
  • display math
  • display math
Process B
  • display math

where [CH3NH2D] and [CH3DNH2] represent the intermediates in process B. Similar processes were proposed for the H–D substitution of solid CH3OH, where CH3OH was deuterated through surface reactions with D atoms at 10 K (Nagaoka et al. 2007). For CH3OH, a combination of H abstraction and the subsequent D-addition reactions explains their experimental results very well (Nagaoka et al. 2007).

Assuming that H–D substitution of CH3NH2 proceeds through process A, the reaction should be initiated by H-atom abstraction from either the methyl or amino groups, as follows:

  • display math(1a)
  • display math(1b)

As we were unable to firmly identify these radicals (CH3NH and CH2NH2) or the singly deuterated methylamines (CH3NHD and CH2DNH2) in the reaction product (Fig. 2), it is still unknown which reactions are dominant in this experiment. Our quantum-chemical calculations show that Reactions 1a and 1b have large activation barriers, Ea, of approximately 36 and approximately 20 kJ mol−1 (Table 2), respectively.

Table 2. Summary of the calculated activation barrier Ea (in kJ mol−1) and the effective rate constant k′ (in min−1) for H (D) abstraction and D (H) addition reactions
ReactionRelated functional groupNumberProcessEa (kJ mol−1)k′ (min−1)Relative rate
  1. a

    The attenuation rates of the column densities of the parent molecules by multiple reactions.

CH3NH2 + D → CH3NH + HDamino1aA36.42.3 ± 0.1 × 10−1a1.00
CH3NH2 + D → CH2NH2 + HDmethyl1bA20.3
CH3NH2 + D → [CH3NH2D]amino1cB, C18.2
[CH3NH2D] → CH3NHD + Damino1dB21.4n.d. 
CH3NH2 + D → [CH3DNH2]methyl1eB157n.d. 
CH3NHD + D → CH3ND + HDamino A36.5n.d. 
CH2DNH2 + D → CHDNH2 + HDmethyl A20.2n.d. 
CHD2NH2 + D → CD2NH2 + HDmethyl A20.3n.d. 
       
CD3NH2 + D → CD3NH + HDamino2aA36.42.9 ± 0.3 × 10−2a0.13
CD3NH2 + D → [CD3NH2D]amino2cB, C18.1
CD3NHD + D → [CD3NHD2]amino3cB, C18.6n.d. 
       
CH3ND2 + D → CH2ND2 + HDmethyl5aA20.22.5 ± 0.1 × 10−11.09
       
CD3ND2 + H → CD3ND + HDamino10aA45.51.2 ± 0.1 × 10−1a0.52
CD3ND2 + H → CD2ND2 + HDmethyl10bA25.1
CD3ND2 + H → [CD3NHD2]amino10cB, C21.4
CH3ND2 + H → CH3ND + HDamino11aA45.43.1 ± 0.3 × 10−2a0.13
CH3ND2 + H → [CH3NHD2]amino11cB, C21.5
CD3NH2 + H → CD2NH2 + HDmethyl13aA25.11.0 ± 0.2 × 10−10.43

In addition to process A, it is likely that the amino group of CH3NH2 is also deuterated through process B. We have found that the energy barrier of H–D substitution in amino groups through process B is competitive with that of Reaction 1a and much smaller than that of Reaction 1b; see Fig. 4. The reaction is initiated by the formation of a stable intermediate, [CH3NH2D],

image

Figure 4. Potential energy diagram of the reactions of methylamine and D atoms calculated with CCSD(T)/aug-cc-pVTZ//B3LYP/6-311G(d,p) level.

Download figure to PowerPoint

  • display math(1c)

The energy barrier of Reaction 1c was calculated at approximately 18 kJ mol−1. The dissociation channel of this intermediate,

  • display math(1d)

is endothermic by 0.5 kJ mol−1 with the barrier of approximately 21 kJ mol−1 (Fig. 4). The energy barrier of H–D substitution in methyl groups through the same process initiated by the following reaction,

  • display math(1e)

was found to be 157 kJ mol−1.

At temperatures as low as 10 K, Arrhenius-type reactions characterized by an activation barrier of >2000 K (approximately 17 kJ mol−1) would not occur. Therefore, regardless of which reaction is more effective, H–D substitution of CH3NH2 via process A should proceed through quantum tunneling on a low-temperature surface. Quantum tunneling is known to be an important process for molecular synthesis on interstellar grains at very low temperatures (Watanabe and Kouchi 2008). As the value of Ea for H-atom abstraction from the methyl group of CH3NH2 by H (D) atoms is much lower than that associated with the amino group (Zhang et al. 2005; Kerkeni and Clary 2007; Table 2), it is reasonable to assume that H–D substitution reactions via process A are faster for the methyl group than for the amino group of solid CH3NH2.

As for H–D substitution in amino group via process B, the initial step (Reaction 1c) may occur through quantum tunneling; however, the second step Reaction 1d does not proceed through quantum tunneling because of endothermic reaction (Fig. 4). Moreover, the heat of Reaction 1c (9.3 kJ mol−1) is not enough to overcome the activation barrier of Reaction 1d (21.4 kJ mol−1). Therefore, CH3NHD would not form through via process B. Instead, reaction of [CH3NH2D] with additional D atoms may be an alternative route to CH3NHD formation:

  • display math(1f)

A combination of intermediate formation and a subsequent reaction with D atoms is denoted as process C. As this is a radical–radical reaction, it does not have an activation barrier. Hence, it should proceed very quickly when the reactants meet on the surface, resulting in deuteration of the amino group. We therefore consider that process C is also an effective route for deuteration of amino groups, just like process A, although it was not possible to determine the degree of each contribution. The intermediate was not identified in the difference spectra of the reaction product (Fig. 2b) probably because Reaction 1f proceeds very quickly. In contrast, Reaction 1e will not proceed at 10 K even by quantum tunneling because of the extremely large barrier of 157 kJ mol−1 (Table 2). To further investigate the H–D substitution reactions of methylamine at the functional group level, we performed additional experiments using methylamine isotopologues CD3NH2 or CH3ND2, instead of CH3NH2.

Figure 5 shows changes in the spectrum of solid CD3NH2 after exposure to D atoms. Peaks derived from ND2 stretch at approximately 2400 cm−1 appeared concurrently with the consumption of CD3NH2, which indicates that CD3ND2 was formed through the following reactions:

image

Figure 5. a) Infrared spectrum of solid CD3NH2 before exposure to H atoms, and b) variations in the difference spectrum after exposure to D atoms for up to 120 min.

Download figure to PowerPoint

  • display math(2a)
  • display math(2b)
  • display math(3a)
  • display math(3b)

Note that D-atom abstraction from the methyl group of CD3NH2 by D atoms may also occur. However, the formed CD2NH2 radical will soon react with another D atom to yield CD3NH2 again, as follows:

  • display math(4a)
  • display math(4b)

As Reaction 4b is a radical–radical reaction, it is assumed to proceed very quickly, even at 10 K. Hence, the column density of CD3NH2 is apparently unchanged through Reactions 4a and 4b.

H–D substitution of the amino group of CD3NH2 may also proceed via process C:

  • display math(2c)
  • display math(2d)
  • display math(3c)
  • display math(3d)

where [CD3NH2D] and [CD3NHD2] are the intermediate species in CD3ND2 formation. The barrier heights for both reactions 2c and 3c were calculated at approximately 18 kJ mol−1 (Table 2), which is almost equivalent to that of Reaction 1c.

As regards the experiment on the H–D substitution of CH3ND2 (Fig. 6), it is clearly shown that sharp peaks derived from CD3 stretch appeared at around 2000–2400 cm−1 after D exposure, in concurrence with the consumption of CH3ND2, indicating the formation of CD3ND2 through the following reactions:

image

Figure 6. a) Infrared spectrum of solid CH3ND2 before exposure to D atoms, and b) variations in the difference spectrum after exposure to D atoms for up to 30 min.

Download figure to PowerPoint

  • display math(5a)
  • display math(5b)
  • display math(6a)
  • display math(6b)
  • display math(7a)
  • display math(7b)

D-atom abstraction from the above methylamine isotopologues may also occur, but it would not affect the changes in the column density of methylamines for the same reasons as explained above.

The rate equation describing the H–D substitution of methylamines (CH3NH2, CD3NH2, and CH3ND2) is defined as

  • display math(8)

where N(MA)t, ND, and kn are the column density of the deposited methylamine isotopologues at time t, the surface density of D atoms, and the rate constant for reaction n, respectively. Integration of Equation 8 gives

  • display math(9)

where N(MA)0 represents the initial column density of methylamine isotopologues and k′n is the effective rate constant (k′n = kn × ND). As it is very difficult to measure the surface density of D atoms ND in the present experimental setup, we can only determine the value of k′n for the H–D substitution of methylamines. Note that the H–D substitution reaction is composed of two steps: H abstraction followed by D addition (process A) or D addition followed by H abstraction (process B). In either process, the second reaction (radical–radical reaction) is much faster than the initial reaction, which has the activation barrier of >18 kJ mol−1 (Table 2). Hence, the effective rate constant for the slower reaction can be approximated as that for the overall H–D substitution reaction. In the case of Reaction 5, for example, the effective rate constant k′5a is approximated as k′5.

Figure 7 shows variations in the relative abundance of solid methylamine isotopologues (CH3NH2, CD3NH2, and CH3ND2) after exposure to D atoms. About 8–15% of parent methylamines were consumed after 2-h exposure to D atoms (Fig. 7). This consumption corresponds to at most the top few MLs when the solids have well-packed flat surfaces. However, it should be noted that as solid methylamines produced by vapor-deposition at 10 K are expected to be amorphous, the surface area would be significantly larger than the flat surface mentioned above. Nevertheless, the present result does not exclude the possibility that D atoms can penetrate into the solid methylamines at 10 K. Future study will clarify diffusion of D atoms into the solid methylamine at low temperatures.

image

Figure 7. Variations in column densities normalized to initial methylamines with D-exposure time. Symbols include statistical errors.

Download figure to PowerPoint

By fitting the plots of the normalized column densities of CH3NH2, CD3NH2, and CH3ND2 against D-exposure time (Fig. 7) using the rate Equation 9, we determined k′1 = (2.3 ± 0.1) × 10−1, k′2 = (2.9 ± 0.3)× 10−2, and k′5a = (2.5 ± 0.1) × 10−1 min−1. As for k′1, an alphabetical subscript was not added, because we cannot distinguish between Reactions 1a, 1b, and 1c in the result obtained (Fig. 2). Likewise, it was not added to k′2 for similar reasons. Instead, k′1 and k′2 represent the attenuation rates of the column densities of CH3NH2, and CD3NH2, respectively. We found that H–D substitution of the methyl group of CH3ND2 proceeds an order of magnitude faster than that of the amino group of CD3NH2 (Table 2).

The value of k′1 is consistent with that of k′5a, implying that H abstraction from the methyl group is the rate-determining step of the H–D substitution of CH3NH2. In general, when the fastest reaction is the rate-determining step under conditions where multiple reactions occur, all possible reactions should proceed in parallel (Lasaga 1981). Hence, the reaction CH3NH2 + D would lead to formation of CD3ND2 through all possible pathways represented by downward arrows in Fig. 8, although the contributions of the individual reactions in Fig. 8 to CD3ND2 formation may differ significantly.

image

Figure 8. Surface-reaction network for the deuteration of CH3NH2 and the hydrogenation of CD3ND2. Solid and dotted arrows represent reactions involving D and H atoms, respectively. The numbers in parentheses represent simplified expressions of methylamine isotopologues CH3–iDiNH2–jDj (= 0–3, = 0–2) as used in our model (see text). Effective rate constants for reactions designated by thick arrows were experimentally determined in the present study. Methylamine isotopologues used in the present study are shown in square frames.

Download figure to PowerPoint

Exposure of Solid Methylamine (CD3ND2, CH3ND2, or CD3NH2) to H Atoms

Figure 9 shows the infrared absorption spectrum of solid CD3ND2 as well as the difference spectra obtained after H exposure to CD3ND2 for up to 30 min. Several peaks appeared in the spectra after exposure to H atoms accompanied by consumption of CD3ND2. The appearance of NH2 stretch, CH3 stretch, and NH2 deformation at approximately 3300, 2800, and 1600 cm−1, respectively, strongly suggests that both methyl and amino groups of CD3ND2 were D–H substituted after its exposure to H atoms. This is supported by the QMS spectrum of the H-exposed CD3ND2, where the signal of mass 31 (CH3NH2) was strongly enhanced after H exposure, mainly because of the formation of CH3NH2 (Fig. 10).

image

Figure 9. a) Infrared spectrum of solid CD3ND2 before exposure to H atoms, and b) variations in the difference spectrum after exposure to H atoms for up to 30 min.

Download figure to PowerPoint

image

Figure 10. TPD spectra (m/z = 31) of solid CD3ND2 before and after exposure to H atoms. A small peak observed for the intact CD3ND2 could be derived from fragment ions of impurities.

Download figure to PowerPoint

The D–H substitution of deuterated methylamine is considered to occur through the following processes: a combination of D-abstraction and H-addition reactions (process A′: counterpart of process A for H-D substitution), formation of intermediate species by H addition followed by elimination of D atoms from the methyl and/or amino groups (process B′), and/or the reaction of the intermediate with H atoms (process C′). On this basis, the D–H substitution of CD3ND2 would be initiated by the following reactions:

  • display math(10a)
  • display math(10b)
  • display math(10c)

The formation of intermediate [CD3HND2] was not expected to occur due to the same reason for [CH3DNH2] formation as explained earlier. As the value of Ea for Reaction 10b is approximately 20 kJ mol−1 lower than that for Reaction 10a (Table 2), Reaction 10b would occur preferentially instead of Reaction 10a. In addition, the barrier height for Reaction 10c is much lower than that of Reaction 10a (Table 2); process C′ would have a larger contribution to the D–H substitution of the amino group. We next performed similar experiments using CH3ND2 and CD3NH2, instead of CD3ND2.

Figures 11 and 12 show changes in the spectra of solid CH3ND2 and CD3NH2, respectively, after exposure to H atoms for up to 120 min. For CH3ND2, NH2 stretching and deformation bands appeared at approximately 3300 and 1600 cm−1, respectively (Fig. 11); for CD3NH2, CH3 stretching and deformation bands appeared at approximately 2800 and approximately 1400 cm−1, respectively (Fig. 12). The appearance of these new peaks strongly indicates that CH3NH2 was formed from CH3ND2 and CD3NH2 through reactions with H atoms, as follows:

image

Figure 11. a) Infrared spectrum of solid CH3ND2 before exposure to H atoms, and b) variations in the difference spectrum after exposure to H atoms for up to 120 min.

Download figure to PowerPoint

image

Figure 12. a) Infrared spectrum of solid CD3NH2 before exposure to H atoms, and b) variations in the difference spectrum after exposure to H atoms for up to 120 min.

Download figure to PowerPoint

(i) for the amino group,

  • display math(11a)
  • display math(11b)
  • display math(11c)
  • display math(11d)
  • display math(12a)
  • display math(12b)
  • display math(12c)
  • display math(12d)

and (ii) for the methyl group,

  • display math(13a)
  • display math(13b)
  • display math(14a)
  • display math(14b)
  • display math(15a)
  • display math(15b)

where [CD3NHD2] and [CH3NHD2] represent the intermediates in CH3NH2 formation. H-atom abstraction from the methyl or amino groups of methylamine isotopologues by H atoms would also occur during exposure to H atoms. However, this process does not affect the variations in the column density of those methylamines. That is, X-atom (X = H or D) abstraction by X atoms followed by the addition of an X atom did not modify the column density of methylamines, as explained in the previous section.

Variations in the normalized column density of CD3ND2, CH3ND2, and CD3NH2 during exposure to H atoms are plotted in Fig. 13. By fitting these plots to the rate Equation 9, we determined the values of k′10, k′11, and k′13a at (1.2 ± 0.1) × 10−1, (3.1 ± 0.3) × 10−2, and (1.0 ± 0.2) × 10−1 min−1, respectively (Table 2). In the case of D–H substitution reactions, the effective rate constant k′n is defined as kn × NH where NH is the surface density of H atoms. The effective rate constants k′11 and k′13a are considered to represent those of the D–H substitution of amino and methyl groups of methylamines, as expected in the H–D substitution of methylamines. The rate of D–H substitution in the amino group of CH3ND2 was three times lower than that in the methyl group of CD3NH2 (Table 2). The value of k′13a was almost the same as that of k′10, indicating that D abstraction from the methyl group is the rate-determining step for D–H abstraction of solid CD3ND2. This is consistent with the H–D substitution of solid methylamine, as explained above. The reaction CD3ND2 + H would therefore lead to the formation of CH3NH2 through all possible pathways represented by the upward arrows in Fig. 8, although the contributions of the individual reactions to CH3NH2 formation may differ.

image

Figure 13. Variations in column densities normalized to initial methylamines with H-exposure time. Symbols include statistical errors.

Download figure to PowerPoint

Possible D/H Ratio of Methylamine in Molecular Clouds from Surface Reactions

The present study experimentally demonstrated that solid CH3NH2 became enriched in D through reactions with D atoms on a solid surface at 10 K, while its deuterated isotopologues were depleted in D when reacting with H atoms under the same conditions. Deuterated methylamines, such as CH2DNH2 and CH3NHD, have never been found in molecular clouds; only CH3NH2 was found in the gas phase (Fourikis et al. 1974; Kaifu et al. 1974; Nummelin et al. 2000). Nevertheless, we believe that the nondetection to date of deuterated methylamines in molecular clouds does not indicate their absence in those environments. Using their effective rate constants (Table 2), we roughly calculated the relative abundance of methylamine isotopologues, which can reach through surface reactions at 10 K. As all reactions shown in Fig. 8 are expected to occur, we define a rate equation for each methylamine isotopologue:

  • display math(16)
  • display math(17)
  • display math(18)
  • display math(19)
  • display math(20)
  • display math(21)
  • display math(22)
  • display math(23)
  • display math(24)
  • display math(25)
  • display math(26)
  • display math(27)

where Nij represents the abundance of methylamine isotopologues CHiD3–iNHjD2–j (= 0–3, = 0–2; e.g., N22 represents the abundance of CH2DNH2) and NX represents the surface density of X atoms, respectively. Effective rate constants for H–D substitution in methyl or amino groups are given as kC(H–D) or kN(H–D), and those for D–H substitution are kC(D–H) or kN(D–H), respectively. Although the present study did not reveal whether or not secondary kinetic isotope effects (SKIEs) on reaction rates are present, we assumed for simplicity that there are no SKIEs for the rates of H–D and D–H substitution reactions (non-SKIEs model). This means, for example, that we assume Reactions 5a, 6a, and 7a to proceed at the same rate.

The relative abundance of each methylamine at time t was obtained by solving the rate equations. Figure 14 shows variations in the relative abundances of methylamine isotopologues with time in molecular clouds under assumptions that the density of H atom is 1 cm−3, the atomic D/H ratio is constant (0.01), and only CH3NH2 is present at = 0. At = 106 years (a typical lifetime of molecular clouds), for example, the fractional abundance of each deuterated methylamine varies significantly, ranging from 7.5 × 10−2 for CH2DNH2 to 6.3 × 10−10 for CD3ND2. This variation clearly indicates that the majority of deuterated methylamines present after 106 years is d1-methylamines (CH2DNH2 and CH3NHD), which account for more than 97% of the total deuterated methylamines. Then, we next show variations in the calculated abundance of d1-methylamines relative to that of CH3NH2 with relevance to the ratio of D and H atoms accreting on grains from the gas phase at = 106 years. It should be noted that the value of N22/N32, that is, [CH2DNH2]/[CH3NH2] is always higher than the atomic D/H by a factor of about 2 (Fig. 15). On the other hand, N31/N32 ([CH3NHD]/[CH3NH2]) is always lower than the atomic D/H; however, it is still more than 50% of the atomic D/H. Recent theoretical models suggest that atomic D/H ratio is at most on the order of 10−2 to 10−1 through the evolution of molecular clouds (Roberts et al. 2002, 2003). Based on the high [CH2DNH2]/[CH3NH2] and [CH3NHD]/[CH3NH2] brought by surface reactions and high atomic D/H in molecular clouds, we predict that d1-methylamines are possibly present in those environments, although their absolute abundances may not be so high. In that case, CH2DNH2 is expected to be the primary deuterated methylamine. We highly desire that future observational studies with better S/N find such deuterated methylamines in molecular clouds.

image

Figure 14. Variations in the fractional abundance of methylamine isotopologues with time under the conditions of atomic D/H = 0.01. nij represents the fractional abundance of methylamine isotopologues CHiD3–iNHjD2–j (= 0–3, = 0–2).

Download figure to PowerPoint

image

Figure 15. Variations in the calculated values of [CH2DNH2]/[CH3NH2] (blue) and [CH3NHD]/[CH3NH2] (red) relative to the ratio of D and H atoms accreting onto interstellar grains from the gas phase at = 106 years. Solid and dashed lines represent those variations for non-SKIEs and SKIEs models, respectively. The dotted line represents a 1:1 relationship.

Download figure to PowerPoint

If SKIEs exist on the hydrogen substitution of methylamine as for CH3OH, the variations in the relative abundance of methylamine isotopologues may differ from those estimated in non-SKIEs model. Variations in [CH2DNH2]/[CH3NH2] and [CH3NHD]/[CH3NH2] with relevance to atomic D/H are shown by dashed lines in Fig. 15 where SKIEs are taken into account for the hydrogen substitution reactions in methyl group (SKIEs model). The degree of SKIEs is followed by Nagaoka et al. (2007) where H abstraction from the methyl group of solid CH3OH at 10 K is about 1.4 and 2 times faster than that from CH2DOH and CHD2OH, respectively. In that case, the abundance of CH2DNH2 is enhanced and that of CH3NHD is suppressed compared with the above non-SKIEs model shown by solid lines in Fig. 15. This result also suggests that CH2DNH2 is the primary D substituted methylamine present in molecular clouds. Further laboratory studies as to SKIEs for the H–D and D–H substitution of methylamines will give a better estimate for the D/H of methylamine by low-temperature surface reactions. Experimental and theoretical studies on the deuterium fractionation of methylamine in the gas phase are also highly desired for the full understanding of possible methylamine D/H in molecular clouds.

Methylamines were used for the formation of glycine and its structural isomer methylcarbamic acid under astrophysically relevant conditions (e.g. Holtom et al. 2005; Bossa et al. 2008; Lee et al. 2009). However, the hydrogen isotopic fractionation during the formation of those important molecules in view of astrobiology has not been studied so far, nor D-substituted methylamines have been used for such experiments. Nevertheless, we expect that reaction products become enriched in D when deuterated methylamines are used for their formation. Future laboratory experiments will clarify D enrichment of glycine and its isomer during their formation from deuterated methylamines.

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental Procedure
  5. Quantum-Chemical Calculation
  6. Results and Discussion
  7. Conclusions
  8. Acknowledgments
  9. Editorial Handling
  10. References

The present study experimentally investigated H–D and D–H substitutions of solid methylamine isotopologues through surface reactions at 10 K. When solid CH3NH2 reacted with D atoms, deuterated methylamine CD3ND2 was formed and when CD3ND2 reacted with H atoms, hydrogenated methylamine CH3NH2 was formed. H–D (D–H) substitution occurred faster in the methyl group of methylamines than in the amino group. It is expected that a series of H (D) abstraction–D (H) addition reactions is the main process for H–D (D–H) substitution in the methyl group. In addition to this substitution–addition process, formation of an intermediate species like [CH3NH2D] followed by the reaction with D (H) atoms may also be effective for the H–D (D–H) substitution in amino groups. Based on the effective rate constants for H–D and D–H substitution reactions experimentally obtained in the present study, we predict that singly deuterated methylamine CH2DNH2 is the most abundant isotopologue formed by surface reactions within the typical lifetime of molecular clouds. The CH2DNH2/CH3NH2 ratio brought by surface reactions is expected to be always higher than atomic D/H ratio in molecular clouds.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental Procedure
  5. Quantum-Chemical Calculation
  6. Results and Discussion
  7. Conclusions
  8. Acknowledgments
  9. Editorial Handling
  10. References

We acknowledge the associate editor Dr. Scott Sandford and two anonymous reviewers for their helpful comments to improve the earlier version of our manuscript. This work was partly supported by a Grant-in-Aid for Scientific Research from the Japan Society for the Promotion of Science (JSPS) and by a research fellowship from JSPS for Young Scientists (Y. Oba).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental Procedure
  5. Quantum-Chemical Calculation
  6. Results and Discussion
  7. Conclusions
  8. Acknowledgments
  9. Editorial Handling
  10. References
  • Bossa J.-B., Borget F., Duvernay F., Theule P., and Chiavassa T. 2008. Formation of neutral methylcarbamic acid (CH3NHCOOH) and methylammonium methylcarbamate [CH3NH3+][CH3NHCO2] at low temperature. Journal of Physical Chemistry A 112:51135120.
  • Bossa J.-B., Duvernay F., Theule P., Borget F., d'Hendecourt L., and Chiavassa T. 2009. Methylammonium methylcarbamate thermal formation in interstellar ice analogs: A glycine salt precursor in protostellar environments. Astronomy & Astrophysics 506:601608.
  • Bossa J.-B., Borget F., Duvernay F., Danger G., Theule P., and Chiavassa T. 2012. Solid-state methylamine VUV irradiation study using carbon monoxide as an H radical scavenger. Australian Journal of Chemistry 65:129137.
  • Ceccarelli C., Castets A., Loinard L., Caux E., and Tielens A. G. G. M. 1998. Detection of doubly deuterated formaldehyde towards the low-luminosity protostar IRAS 16293–2422. Astronomy & Astrophysics 338:L43L46.
  • Dunning T. H., Jr. 1989. Gaussian basis sets for use in correlated molecular calculations. I. The atoms boron through neon and hydrogen. The Journal of Chemical Physics 90:10071023.
  • Durig J. R. and Zheng C. 2001. Infrared spectra of krypton solutions of methylamine. Structural Chemistry 12:137148.
  • Durig J. R., Bush S. F., and Baglin F. G. 1968. Infrared and raman investigation of condensed phases of methylamine and its deuterium derivatives. The Journal of Chemical Physics 49:21062117.
  • Elsila J. E., Glavin D. P., and Dworkin J. P. 2009. Cometary glycine detected in samples returned by Stardust. Meteoritics & Planetary Science 44:13231330.
  • Fourikis N., Takagi K., and Morimoto M. 1974. Detection of interstellar methylamine by its 202 → 110 Aa-state transition. The Astrophysical Journal 191:L139L141.
  • Fourikis N., Takagi K., and Saito S. 1977. Probable detection of interstellar methylamine-d (CH3NHD). The Astrophysical Journal 212:L33L37.
  • Frisch M. J, Trucks G. W., Schlegel H. B., Scuseria G. E., Robb M. A., Cheeseman J. R., Montgomery Jr. J. A., Vreven T., Kudin K. N., Burant J. C., Millam J. M., Iyengar S. S., Tomasi J., Barone V., Mennucci B., Cossi M., Scalmani G., Rega N., Petersson G. A., Nakatsuji H., Hada M., Ehara M., Toyota K., Fukuda R., Hasegawa J., Ishida M., Nakajima T., Honda Y., Kitao O., Nakai H., Klene M., Li X., Knox J. E., Hratchian H. P., Cross J. B., Bakken V., Adamo C., Jaramillo J., Gomperts R., Stratmann R. E., Yazyev O., Austin A. J., Cammi R., Pomelli C., Ochterski J. W., Ayala P. Y., Morokuma K., Voth G. A., Salvador P., Dannenberg J. J., Zakrzewski V. G., Dapprich S., Daniels A. D., Strain M. C., Farkas O., Malick D. K., Rabuck A. D., Raghavachari K., Foresman J. B., Ortiz J. V., Cui Q., Baboul A. G., Clifford S., Cioslowski J., Stefanov B. B., Liu G., Liashenko A., Piskorz P., Komaromi I., Martin R. L., Fox D. J., Keith T., Al-Laham M. A., Peng C. Y., Nanayakkara A., Challacombe M., Gill P. M. W., Johnson B., Chen W., Wong M. W., Gonzalez C., and Pople J. A. 2004. Gaussian 03, Revision D.02. Wallingford, CT: Gaussian, Inc.
  • Gardner E. P. and McNesby J. R. 1980. Methylamine formation in the vacuum ultraviolet photolysis of methane and ammonia mixtures. Journal of Photochemistry 13:353356.
  • Gerlich D. and Schlemmer S. 2002. Deuterium fractionation in gas-phase reactions measured in the laboratory. Planetary and Space Science 50:12871297.
  • Glavin D. P., Dworkin J. P., and Sandford S. A. 2008. Detection of cometary amines in samples returned by Stardust. Meteortics & Planetary Science 43:399413.
  • Herbst E. 1985. The rate of the radiative association reaction between CH3+ and NH3 and its implications for interstellar chemistry. The Astrophysical Journal 292:484486.
  • Hidaka H., Kouchi A., and Watanabe N. 2007. Temperature, composition, and hydrogen isotope effect in the hydrogenation of CO on amorphous ice surface at 10–20 K. The Journal of Chemical Physics 126:204707.
  • Hidaka H., Watanabe M., Kouchi A., and Watanabe N. 2009. Reaction routes in the CO–H2CO–dn–CH3OH-dm system clarified from H(D) exposure of solid formaldehyde at low temperatures. The Astrophysical Journal 702:291300.
  • Holtom P. D., Bennett C. J., Osamura Y., Mason N. J., and Kaiser R. I. 2005. A combined experimental and theoretical study on the formation of the amino acid glycine (NH2CH2COOH) and its isomer (CH3NHCOOH) in extraterrestrial ices. The Astrophysical Journal 626:940952.
  • Kaifu N., Morimoto M., Nagane K., Akabane K., Iguchi T., and Takagi K. 1974. Detection of interstellar methylamine. The Astrophysical Journal 191:L135L137.
  • Kerkeni B. and Clary D. C. 2007. Quantum scattering study of the abstraction reactions of H atoms from CH3NH2. Chemical Physics Letters 438:17.
  • Lasaga A. C. 1981. Rate laws and chemical reactions. In Kinetics of geochemical processes, edited by Lasaga A. C. and Kirkpatrick R. J. Reviews in Mineralogy, vol. 8. Washington, D.C.: Mineralogical Society of America. pp. 168.
  • Lee C.-W., Kim J.-K., Moon E.-S., Minh Y. C., and Kang H. 2009. Formation of glycine on ultraviolet-irradiated interstellar ice-analog films and implications for interstellar amino acids. The Astrophysical Journal 697:428435.
  • Leung C. M., Herbst E., and Huebner W. F. 1984. Synthesis of complex molecules in dense interstellar clouds via gas-phase chemistry: A pseudo time-dependent calculation. The Astrophysical Journal Supplement Series 56:231256.
  • MacLeod J. M., Avery L. W., Broten N. W., and Oka T. 1979. Failure to confirm the existence of interstellar methylamine-d (CH3NHD). The Astrophysical Journal 231:401403.
  • Maeda S. and Ohno K. 2006. Generation mechanisms of amino acids in interstellar space via reactions between closed-shell species: Significance of higher energy isomers in molecular evolution. The Astrophysical Journal 640:823828.
  • Moon E.-S., Lee C.-W., Kim J.-K., Park S.-C., and Kang H. 2008. UV-induced protonation of molecules adsorbed on ice surfaces at low temperature. The Journal of Chemical Physics 12:191101.
  • Nagaoka A., Watanabe N., and Kouchi A. 2005. H-D substitution in interstellar solid methanol: A key route for D enrichment. The Astrophysical Journal 624:L29L32.
  • Nagaoka A., Watanabe N., and Kouchi A. 2007. Effective rate constants for the surface reaction between solid methanol and deuterium atoms at 10 K. Journal of Physical Chemistry A 111:30163028.
  • Nummelin A., Bergman P., Hjalmarson Å., Friberg P., Irvine W. M., Ohishi M., and Saito S. 2000. A three-position spectral line survey of Sagittarius B2 between 218 and 263 GHz. II. Data analysis. The Astrophysical Journal Supplement Series 128:213243.
  • Oba Y., Watanabe N., Hama T., Kuwahata K., Hidaka H., and Kouchi A. 2012. Water formation through a quantum tunneling surface reaction, OH + H2, at 10 K. The Astrophysical Journal 749:67.
  • Ogura K., Migita C. T., and Yamada T. 1988. Photochemical formation of methylamine and ethylenediamine from gas mixtures of methane, ammonia, and water. Chemistry Letters 1988:15631566.
  • Osamura Y., Roberts H., and Herbst E. 2005. The gas-phase deuterium fractionation of formaldehyde. The Astrophysical Journal 621:348358.
  • Parise B., Castets A., Herbst E., Caux E., Ceccarelli C., Mukhopadhyay I., and Tielens A. G. G. M. 2004. First detection of triply-deuterated methanol. Astronomy & Astrophysics 416:159163.
  • Roberts H. and Miller T. J. 2000. Gas-phase formation of doubly-deuterated species. Astronomy & Astrophysics 364:780784.
  • Roberts H., Herbst E., and Millar T. J. 2002. The importance of new rate coefficients for deuterium fractionation reactions in interstellar chemistry. Monthly Notices of the Royal Astronomical Society 336:283290.
  • Roberts H., Herbst E., and Millar T. J. 2003. Enhanced deuterium fractionation in dense interstellar cores resulting from multiply deuterated H3+. The Astrophysical Journal 591:L41L44.
  • Stantcheva T. and Herbst E. 2003. Deuterium fractionation on interstellar grains studied with the direct master equation approach. Monthly Notices of the Royal Astronomical Society 340:983988.
  • Takagi K. and Kojima T. 1971. Microwave spectrum of methylamine. Journal of the Physical Society of Japan 30:11451157.
  • Tamagake K. and Tsuboi M. 1974. Inversion and internal rotation in methyl-d-amine: Microwave spectrum. Journal of Molecular Spectroscopy 53:204220.
  • Taquet V., Ceccarelli C., and Kahane C. 2012. Formaldehyde and methanol deuteration in protostars: Fossils from a past fast high-density pre-collapse phase. The Astrophysical Journal 748:L3.
  • Theule P., Borget F., Mispelaer F., Danger G., Duvernay F., Guillemin J. C., and Chiavassa T. 2011. Hydrogenation of solid hydrogen cyanide HCN and methanimine CH2NH at low temperature. Astronomy & Astrophysics 534:A64.
  • Watanabe N. and Kouchi A. 2008. Ice surface reactions: A key to chemical evolution in space. Progress in Surface Science 83:439489.
  • Watanabe N., Nagaoka A., Hidaka H., Shiraki T., Chigai T., and Kouchi A. 2006. Dependence of the effective rate constants for the hydrogenation of CO on the temperature and composition of the surface. Planetary and Space Science 54:11071114.
  • Woon D. E. 2002. Pathways to glycine and other amino acids in ultraviolet-irradiated astrophysical ices determined via quantum chemical modeling. The Astrophysical Journal 571:L177L180.
  • Zhang Q., Zhang R. Q., Chan K. S., and Bello I. 2005. Ab initio and variational transition state approach to β-C3N4 formation: Kinetics for the reaction of CH3NH2 with H. Journal of Physical Chemistry A 109:91129117.