- Email: [email protected]
Influence of methane concentration on the optical indices of Titan’s aerosols analogues
Icarus 221 (2012) 670–677
Contents lists available at SciVerse ScienceDirect
Icarus journal homepage: www.elsevier.com/locate/icarus
Influence of methane concentration on the optical indices of Titan’s aerosols analogues A. Mahjoub ⇑, N. Carrasco, P.-R. Dahoo, T. Gautier, C. Szopa, G. Cernogora Université Versailles St-Quentin, UPMC Univ. Paris 06, CNRS, LATMOS, 11 Blvd. d’Alembert, 78280 Guyancourt, France
a r t i c l e
i n f o
Article history: Received 17 February 2012 Revised 16 August 2012 Accepted 16 August 2012 Available online 11 September 2012 Keywords: Titan Photometry Radiative transfer Titan, atmosphere
a b s t r a c t This work deals with the optical constant characterization of Titan aerosol analogues or ‘‘tholins’’ produced with the PAMPRE experimental setup and deposited as thin films onto a silicon substrate. Tholins were produced in different N2–CH4 gaseous mixtures to study the effect of the initial methane concentration on their optical constants. The real (n) and imaginary (k) parts of the complex refractive index were determined using the spectroscopic ellipsometry technique in the 370–1000 nm wavelength range. We found that optical constants depend strongly on the methane concentrations of the gas phase in which tholins are produced: imaginary optical index (k) decreases with initial CH4 concentration from 2.3 102 down to 2.7 103 at 1000 nm wavelength, while the real optical index (n) increases from 1.48 up to 1.58 at 1000 nm wavelength. The larger absorption in the visible range of tholins produced at lower methane percentage is explained by an increase of the secondary and primary amines signature in the mid-IR absorption. Comparison with results of other tholins and data from Titan observations are presented. We found an agreement between our values obtained with 10% methane concentration, and Imanaka et al. (Imanaka, H., Khare, B.N., Elsila, J.E., Bakes, E.L.O., McKay, C.P., Cruikshank, D.P., Sugita, S., Matsui, T., Zare, R.N. [2004]. Icarus, 168, 344–366) values, in spite of the difference in the analytical method. This confirms a reliability of the optical properties of tholins prepared with various setups but with similar plasma conditions. Our comparison with Titan’s observations also raises a possible inconsistency between the mid-IR aerosol signature by VIMS and CIRS Cassini instruments and the visible Huygens-DISR derived data. The mid-IR VIMS and CIRS signatures are in agreement with an aerosol dominated by an aliphatic carbon content, whereas the important visible absorption derived from the DISR measurement seems to be incompatible with such an important aliphatic content, but more compatible with an amine-rich aerosol. Ó 2012 Elsevier Inc. All rights reserved.
1. Introduction Titan, the biggest Saturn’s moon, has an important astrobiological interest because of its dense atmosphere unique among all Solar System satellites. Titan’s atmosphere is mainly constituted of nitrogen and methane. The energetic particles coming from Saturn’s magnetosphere and the solar UV radiation gives enough energy to dissociate both nitrogen and methane molecules, allowing a complex chemistry to take place (Waite et al., 2007). This complex photochemistry leads to the formation of solid aerosols as a permanent haze around the satellite. Titan’s aerosols play an important role in the climate, the composition and the properties of Titan’s atmosphere as well as its surface. This haze plays an important role in the radiative transfer in the atmosphere of Titan (Samuelson, 1983) and strongly influences its thermal structure (McKay et al., 1989; Rannou et al., 2010). Their optical and physical ⇑ Corresponding author. E-mail address: [email protected] (A. Mahjoub). 0019-1035/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.icarus.2012.08.015
properties are essential to analyze the observational data acquired by the Cassini Orbiter and Huygens probe instruments, among others. Models developed to explain the optical properties of Titan’s atmosphere require knowledge of optical indices of Titan’s haze (see for instance Tomasko et al., 2008 or Rannou et al., 2010). For these reasons, determining the optical properties of the haze is critical in the analysis of the observational data. As no return sample is planned from Titan’s atmosphere in the short term, the use of laboratory experiments is of primary importance to produce analogues of Titan’s aerosols. Indeed, the study of these analogues and of their production mechanisms enables one to provide important information on the processes taking place in Titan’s atmosphere. Several experimental devices have been used as photochemical reactors or plasma discharges. The latter are the most efficient processes to produce solid material named tholins. Several discharges have been used up to now: Direct Current discharges (DC) (Khare et al., 1984), barrier discharges (Horvath et al., 2010), corona discharges (Horvath et al., 2009), Radio frequency discharges
A. Mahjoub et al. / Icarus 221 (2012) 670–677
Inductively coupled (Imanaka et al., 2004) or capacitively coupled as what was used in the present work on the PAMPRE reactor (Szopa et al., 2006). In the PAMPRE reactor tholins are produced as powder (Szopa et al., 2006) or as thin films on substrates (Quirico et al., 2008; Sciamma-O’Brien et al., 2012). Some properties of produced tholins have been studied: particles size and morphology (Hadamcik et al., 2009), production efficiency and elemental analysis (Sciamma-O’Brien et al., 2010), infrared spectroscopy (Quirico et al., 2008; Gautier et al., 2012), solubility (Carrasco et al., 2009), solid NMR (Derenne et al., 2012) and mass spectrometry (Pernot et al., 2010; Carrasco et al., 2009). In Titan’s upper atmosphere (ionosphere) where aerosols are also produced, the CH4 varies from 2% to 10% (Hébrard et al., 2007). Thus, it is of utmost importance to determine the optical constants of tholins synthesized from N2–CH4 gas mixtures with different CH4 concentrations. Moreover, Atreya et al. (2006) calculate that without methane emission source, photochemistry would lead to the irreversible loss of methane from Titan’s atmosphere within 10–100 myr range. A methane emission source is therefore proposed by Tobie et al. (2006) as episodic outgassing processes. Such an irregular source involves a varying atmospheric methane concentration during the evolution of Titan’s atmosphere. The atmospheric aerosols would therefore have been produced in different methane conditions through time. We suspect that the aerosol optical indices are sensitive to this parameter and could imply a change of the optical opacity of the atmosphere. Studying the influence of a varying methane percentage on the aerosol optical indices is thus crucial to understand the evolutionary history of Titan and its atmosphere. Many of plasma discharges producing tholins have been working with an initial 10:90 CH4:N2 mixture (see for instance Khare et al., 1984, or Imanaka et al., 2004). Other works have been made with other initial methane amounts (Khare et al., 1984; Imanaka et al., 2004; Tran et al., 2003; Ramirez et al., 2002; Vuitton et al., 2009). Nevertheless, these experiments have been performed with different experimental setups and the optical indices have often been measured with different analytical techniques. To our knowledge, there is no consistent study of the evolution of the optical properties with the methane concentration. In the present work, tholins are produced in the PAMPRE plasma discharge device. Sciamma-O’Brien et al. (2010) have shown that a stationary state (continuous working conditions) is reached within about 2 min after the beginning of the plasma discharge. In the stationary state the amount of CH4 is approximately half of the injected one, during the film growth (for a RF power of 30 W, 55 sccm gas flow at a pressure of 0.9 mbar). It was also shown that the elemental composition of tholins (powder) changes with the increase of the initial amount of CH4 injected: C content remains constant at a value about 30% whereas N amount decreases from 30% down to 15% and the H amount increases from 30% up to 45%. So a change of the optical constants as a function of initial CH4 concentration is expected. Optical properties of tholins films have previously been studied for the injected gas mixture CH4:N2 = 5:95 (Sciamma-O’Brien et al., 2012). The question of the effect of tholins production conditions on their optical properties remains open. Many studies have been carried out to determine the real and imaginary parts of the refractive index of tholins produced by various experimental techniques (photochemistry, electric discharges). These indices have been determined using different methods: ellipsometry, spectrophotometric measurements (Khare et al., 1984; Imanaka et al., 2004; Tran et al., 2003; Ramirez et al., 2002; Vuitton et al., 2009; Sciamma-O’Brien et al., 2012), and recently cavity ring down aerosol extinction spectroscopy (Hasenkopf et al., 2010). Sciamma-O’Brien et al. (2012) shows that the optical constants obtained with such different techniques are
671
quite different. In fact, the wide variety of initial gas mixtures used in these experiments produced tholins with different elemental composition which are therefore expected to have different optical properties. In addition, different techniques of refractive index measurement and different experimental conditions (e.g. pressure or temperature) can influence the resulting tholins optical properties. In this paper we determine the imaginary and real parts of the complex optical index of tholins produced with four different CH4 concentrations under the same experimental conditions. Tholins are deposited as thin films onto the same substrate: silicon (Si). Optical properties of the four samples are determined using ellipsometry technique. This study allows the investigation of the influence of the methane concentration independently of substrate and optical diagnostic technique. 2. Experimental methodology 2.1. Tholins film production PAMPRE is an experimental cold plasma discharge described in details in previous publications (Szopa et al., 2006; Alcouffe et al., 2010). Briefly a Capacitively Coupled Plasma radio frequency (CCP RF) discharge at 13.56 MHz is driven between a polarized electrode and a grounded cylindrical grid box confining the plasma. The injected CH4:N2 gas mixture can be changed from 0:100 to 10:90. PAMPRE experimental setup allows the production of tholins in volume (tholins grow in suspension in the gas phase) as well as deposited as thin films onto solid substrates. The continuous gas flow rate gives a constant pressure of 0.9 mbar in the reactor during the tholin production. The reactive gas mixture is at room temperature. In the present work, we focused on the study of optical properties of tholins films deposited on substrates as a function of the initial gas mixture injected in the plasma reactor. For this purpose a thin tholins film was deposited onto a 0.5 mm thick and 12 mm diameter polished silicon substrate. We have chosen silicon substrates because silicon is often used for microelectronic devices and its optical properties are well established. The substrate was placed at the center of the grounded electrode inside the plasma confining grid box. The experiment was run for 2 h in order to produce a film no more than 1 lm thick and therefore limit the optical absorption. After tholins film deposition, samples are studied by spectroscopic ellipsometry at atmospheric pressure in order to determine the real and imaginary parts of their refractive index. 2.2. Spectroscopic ellipsometry Spectroscopic ellipsometry is an optical technique based on the measurement of the change in polarized light upon light reflection on a sample (Azzam and Bashara, 1989; Fujiwara, 2007). Because of recent advances in computer technology, the spectroscopic ellipsometry technique has developed rapidly. As a consequence, ellipsometry is actually used routinely to characterize thin films in electronic and optical industry. In spectroscopic ellipsometry, linearly polarized light is incident at an oblique angle onto the sample. The reflected light from the sample surface is, in general, elliptically polarized. The measured quantity in ellipsometry is the so called complex reflectance ratio, expressed as
q¼
rp ¼ tanðwÞeiD rs
ð1Þ
where rp and rs are the complex Fresnel reflection coefficients for p- and s-polarized light, respectively; tan(w) is the ratio of the
672
A. Mahjoub et al. / Icarus 221 (2012) 670–677
amplitude changes and D is the relative difference in phase changes for the p- and s-polarized light components upon reflection on the sample surface. From experimentally determined q, information can be obtained about the optical properties of the material under study in terms of number of layers, their thicknesses, their roughness and the optical constants of each layer at each wavelength. Spectroscopic ellipsometry is an inverse problem, where parameters w and D are measured but the sample optical properties must be obtained by a model (Dahoo et al., 2004; Franta et al., 2005; Loutete-Dangui et al., 2007). Model-generated ellipsometric parameters (w and D) can then be compared to experimental data. The sample optical properties (fit parameters) whose response best matches the experimental data are found using a regression analysis. The mean squared error (MSE) and comparison between the model-generated curves and experimental data are used to estimate the fit quality of the final model. In the study presented here, we used an M-2000V spectroscopic ellipsometer from J.A. Woollam Co. The M-2000V is a rotating compensator ellipsometer with CCD camera to measure all wavelengths simultaneously. Data are collected 20 times per second. To optimize the signal-to-noise ratio, we chose a 100 s long measurement duration for each full spectrum. The light source is a tungsten lamp which covers a wide spectral range from 370 nm to 1000 nm. Within this spectral range, spectroscopic ellipsometry data are collected over 400 points. We used an angle of incidence of 70°. At that angle, the light spot reflected from the tholins film was approximately 8 mm 3 mm, i.e. smaller than the substrate diameter. The spectroscopic ellipsometry measurements were then analyzed using Woollam’s Complete-EASE™ analysis software (Complete-EASE™ Data Analysis Manual By J.A. Woollam Co Inc. June 15, 2008). With this software, it is possible to describe the different layers composing a film using a specific model for each layer. In the Complete-EASE™ software, morphologic characteristics are also considered in the fitting calculation: surface roughness, thickness non-uniformity as well as the angle offset added to the nominal angle of incidence to compensate for small angle alignment errors. The surface roughness is determined by optimizing a mixed external layer composed of air and tholins. The thickness non-uniformity is related to the polarization variation on the whole ellipsometric spot.
For the Si substrate optical indices are well known (see for instance Herzinger et al., 1998), and tabulated data available in the CompleteEASE™ software have been used. As optical properties of tholins are unknown, a parameteric model must be employed. Hadamcik et al. (2009) showed that tholins produced with PAMPRE setup have an amorphous structure. Then, in order to describe tholins film, we have used a Tauc–Lorentz model which is well adapted for absorbing amorphous organic films (Jellison and Modine, 1996). In the Tauc–Lorentz model the imaginary part of the complex relative dielectric constant e = e1 + ie2 is given by:
e2 ¼
AE0 CðE Eg Þ2 2
ðE
E20 Þ2
1 ; for E > Eg þC E E
ð2Þ
2 2
e2 ¼ 0; for E < Eg This expression has four energy parameters: the band gap Eg, the peak in the joint density of states E0, the broadening parameter C, and the prefactor A, which includes the optical transition matrix elements (Jellison and Modine, 1996). The real part of the dielectric function is derived from e2 using the Kramers–Kronig relations (Kramers, 1927; Kronig, 1926):
2
e1 ðxÞ ¼ 1 þ P p
Z
1
0
x0 e2 ðx0 Þ 0 dx x02 x2
ð3Þ
where P is the Cauchy principal value of the integral:
P
Z 0
1
Z f ðx0 Þdx0 ¼ lim d!0
0
xd
f ðx0 Þdx0 þ
Z
1
f ðx0 Þdx0
ð4Þ
xþd
The complex refractive indices are then deduced from the relative dielectric constants by the relation:
e ¼ e1 þ ie2 ¼ ðn þ ikÞ2
ð5Þ
In a previous work, Sciamma-O’Brien et al. (2012) studied similar tholins films produced in PAMPRE setup with a 5:95 CH4:N2 gas mixture deposited onto a Al:SiO2 substrate. In this case, the organic layer was modeled using a mathematical fitting named B-Spline model. The question of the influence of the chosen model on the results has therefore to be addressed. 3.2. Comparison between B-Spline and Tauc–Lorentz fitting models
3. Ellipsometry data analysis 3.1. Data analyzing procedure Samples studied in this work are thin films deposited on Si substrates. The measured ellipsometric parameters w and D result from the interaction of the light with both tholins film and Si substrate. Moreover the interface between film and the ambient air presents some roughness. Then the sample must be treated with a multilayer model, where the interface is considered as an additional layer composed by a mixture of organic material and air. In this work, a model consisting of three layers has been defined (Fig. 1).
Fig. 1. Schematic representation of a layered optical model with tholins film.
Both models, B-Spline and Tauc–Lorentz, can be used to characterize optical properties of absorbing materials (Weber et al., 2009). In order to evaluate the influence of the chosen model on the optical results we have used the w and D parameters recorded by Sciamma-O’Brien et al. (2012) and calculated both thickness and complex index of the tholins films using a Tauc–Lorentz model. With the Tauc–Lorentz model, we obtain a film thickness of 420 ± 15 nm in agreement with the B-Spline model (Sciamma-O’Brien et al., 2012). Fig. 2 shows in linear scales both real and imaginary parts of the optical index obtained by Tauc–Lorentz and B-Spline models. The values obtained with both models are in good agreement in the 370–900 nm range. Sciamma-O’Brien et al. (2012) point-out a validity limit of the B-Spline model, which prevents the k determination in the quasi-transparent region above 900 nm. In fact, as Tauc–Lorentz model is based on a physical description of the dielectric constants, no numerical divergence appears above 900 nm. This makes possible the determination of optical indices in the range 900–1000 nm (see Fig. 2). Moreover the Tauc–Lorentz model can give useful information about the electronic transition of organic materials constituting tholins. So, for the following, the Tauc–Lorentz model has been chosen for tholins films description.
673
A. Mahjoub et al. / Icarus 221 (2012) 670–677
0.018
1.8
Tauc-Lorentz model Bspline model
Tauc-Lorentz model Bspline model
1.7
n
k
0.009
0.000
1.6 1.5
400
600
800
1000
1.4
Wavelength (nm)
600
900
Wavelength (nm)
Fig. 2. Comparison between imaginary (left) and real (right) index obtained by B-Spline and Tauc–Lorentz fitting models for the tholins film layer on AlSiO2 produced in Sciamma-O’Brien et al. (2012).
Fig. 3. Comparison between experimental and model generated w and D spectroscopic ellipsometry parameters recorded for TF5.
3.3. Effect of methane concentration on tholins optical properties Ellipsometric measurements recorded for samples produced in different methane concentrations (1%, 2%, 5% and 10%) are analyzed using the model described in Section 3.1. We subsequently used the following convention to refer to the different tholins films: TFx, x corresponding to the CH4 percentage. As an example, we will discuss hereafter the fitting quality obtained for the TF5 sample. Fig. 3 shows the comparison between experimental D and w parameters and those generated by the model described in Section 3.1. Although many parameters are taken into account by the Complete-EASE™ software (model for each layer, non-uniformity, roughness, etc.), a global fitting quality is quantified by the mean square error (MSE). The agreement between the fits and the experimental measurements is validated by a MSE value of 4 103 to 1.4 102. In our case the MSE value calculated by the Complete-EASE™ software is about 5 103 confirming a good agreement between the fits and the experimental measurements. Table 1 summarizes the results of geometrical properties of the films, the accuracy of the fit and the gap energy. The uncertainty of
the film thickness is given by the non-uniformity calculated by Complete-EASE™ software. On Table 1 are also reported the uncertainties on all the Tauc– Lorentz parameters evaluated by the Complete-EASE™ software. Table 1 shows that the major contributor to the error is by far the uncertainty in the film roughness (between 5% and 9%). We note that the uncertainty on all fitting parameters is less than 9%, this indicates that the model used to determine n and k values is a good representation of studied samples. We note that the thickness of the thin tholins films increases with the injected amount of CH4. For powder, Sciamma-O’Brien et al. (2010) have found a maximum of mass production for 4% of injected CH4 for the same pressure condition as the present work. Therefore the film thickness does not follow the same evolution with the methane concentration as the powder mass production into the gas volume. This difference is not explained yet, but is beyond the present study. If we compare the film thickness (420 ± 15 nm) obtained by Sciamma-O’Brien et al. (2012) for deposition onto commercial mirrors and those of present films deposited on Si substrate (850 ± 30 nm) for the same experimental conditions and deposition duration, we observe than the deposition on Si is more efficient by a factor of 2. This effect is probably due to the electric properties of the substrates and the polarization on the electrodes. In the PAMPRE CCP discharge, the driven electrode reaches a negative self-bias potential (Alcouffe et al., 2010) when the electrode where the substrates are displayed is grounded. Then, the grounded electrode receives a flux of negative species: electrons and negative ions. Mirrors used by Sciamma-O’Brien et al. (2012) are isolating material, when Si substrates used in the present work are semi conductive. As a consequence the flux of negative ions on Si is more intense. As negative ions are important in dusty plasma chemistry (Berndt et al., 2009) this could explain why tholins films are thicker on Si than on mirrors. Fig. 4 shows the n values obtained for different TFx samples as function of wavelength. We note that when CH4 concentration
Table 1 Results given by the model for different TFx (corresponding to the CH4 percentage) samples: geometrical properties of the films, the accuracy of the fit and gap energy of tholins films. Sample
Thickness (nm)
Roughness (nm)
MSE
Eg (eV)
A (eV)
E0
C
TF1 TF2 TF5 TF10
628 ± 15 770 ± 25 850 ± 30 910 ± 40
24 ± 2.0 34 ± 2.2 15 ± 0.8 10 ± 0.9
4.5 103 5.2 103 6.2 103 6 103
2.05 ± 0.15 2.11 ± 0.13 2.13 ± 0.15 2.26 ± 0.14
40.1 ± 0.84 40.7 ± 1.04 40.5 ± 0.93 39.73 ± 0.98
12.4 ± 0.12 13.37 ± 0.13 13.9 ± 0.13 15 ± 0.15
24.4 ± 0.5 24.4 ± 0.5 24.3 ± 0.5 24.3 ± 0.5
674
A. Mahjoub et al. / Icarus 221 (2012) 670–677
1.70 1.68
TF10
1.66
TF5
1.64
TF2
1.62
TF1
bright yellow for 10% of CH4 (Fig. 6). Otherwise, we note an increase of the optical band gap Eg (Table 1) for increasing methane concentration. The quantity Eg correlates with the fraction of sp3bonded carbon atoms, with higher values of Eg corresponding to higher values of the sp3 fraction (see for instance Jellison et al. (1996)). This result is an agreement with IR measurements presented in Section 4. The averaged n and k values from 370 nm to 900 nm by 10 nm increments are given in the Appendix A. We note that any quantitative determination of uncertainty is reported in Appendix A.
n
1.60 1.58 1.56 1.54 1.52 1.50
4. About the relationship between CH4 concentration and the chemical composition of thin films
1.48 1.46 400
500
600
700
800
900
1000
Wavelength (nm) Fig. 4. Real part, n, of the tholins thin films refractive indices produced with different initial methane concentration.
TF1 TF2 TF5
0,1
k
TF10
0,01
400
500
600
700
800
900
1000
Wavelength Fig. 5. Imaginary part, k, of the tholins thin films refractive indices produced with different initial methane concentration.
Results obtained in this paper in terms of n and k values in different CH4 percentages are not self-sufficient to correlate the tholins composition to the optical indices without any information about the chemical composition of these organic materials. To strengthen our discussion on the relationship between the composition of tholins and its optical constants, Mid-Infrared spectra were recorded for tholins produced for the different studied methane concentrations. This Infra Red spectroscopic study is very helpful to describe the evolution of functional groups existing in tholins in order to connect it to absorption in the UV–Visible wavelengths region. Spectra were recorded at the SMIS (Spectroscopy and Microscopy in the Infra Red using Synchrotron) beam line of SOLEIL synchrotron radiation facility, France (Dumas et al., 2006). We used a NicPlan microscope coupled to a Nicolet Magna System 560 Fourier Transformed Infra Red (FTIR) spectrometer with a spectral resolution of 4 cm1. The Infra Red sources used for the present work were the synchrotron line for Mid-Infrared. The detector used was the Mercury–Cadmium–Telluride (MCT) detector of the microscope (Mid IR). Analysis were performed in transmission mode on thin tholins films deposited onto MirrIR substrates (Low-e microscope slides from Kevley Technologies) substrates. For each sample three spectra were collected and blanks were performed on a blank substrate. Using this experimental protocol we have measured the transmitted Infra Red intensity It given by:
It ¼ I0 e2ad
CH 4 % Fig. 6. Colors of powder tholins produced in different CH4 concentrations. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
increases, the real optical index increases from 1.48 for TF1 to 1.58 for TF10 at 1000 nm. The real indices so obtained have reasonable values compared to other measurements of optical indices of tholins films (Sciamma-O’Brien et al., 2012). Fig. 5 shows the values of the imaginary parts of the optical indices obtained for tholins films produced at different percentages of injected methane. On these spectra we observe that k values decrease when the concentration of CH4 increases in agreement with the anti-correlated evolution of the n value: for example at 1000 nm the difference in k value is one order of magnitude between TF1 and TF10 (2.3 102 versus 2.7 103). These results are correlated to the colors of tholins produced in different CH4 concentration: the color varies from dark brown at 1% of CH4 to
ð6Þ
where I0 is the incident beam intensity, a is the absorption coefficient and d is the film thickness. The factor 2 in the equation is due to the transmission measurement method: the beam crosses the tholins film two times. The film thickness d is determined by spectroscopic ellipsometry on equivalent substrates (for more details see Gautier et al., 2012). The values are respectively 440 nm; 490 nm; 550 nm and 580 nm, for CH4 amounts of 1%; 2%; 5% and 10%. Fig. 7 shows the absolute comparison of the linear absorption coefficient a in the 2500–4000 cm1 range as a function of the initial methane concentration. In all these spectra we observe a broad doublet centered at 3205 and 3325 cm1 respectively. These bands are attributed to primary and secondary amines. A system of three bands appears at 2871, 2930 and 2970 cm1 these bands being attributed to CH2 and CH3 stretching modes. Comparison of spectra revealed a dramatic dependence of the intensities of different bands as function of the injected methane percentage. Intensities of NH and NH2 bands decrease as methane concentration increases on one hand and intensities of CH2 and CH3 bands increase with methane concentration on the other. These observations are very helpful to understand the dependence of tholins absorption in the UV–Visible range. Indeed, the very broad absorption centered around 220 nm due to the n ? r⁄ transition of amines groups is known to be relatively intense from the
675
A. Mahjoub et al. / Icarus 221 (2012) 670–677
0.10
NH,NH2
3000
1% 2% 5% 10 %
2500 CH2, CH3
0.08
0.06
1500
K
-1
α (cm )
2000
this work Khare et al. (1984) (P=0.2 mbar) Imanaka et al. (2004) (P=1.6 mbar) Imanaka et al. (2004)(P= 0.64 mbar)
0.04
1000 500
0.02
0 0.00
-500 2400 2600 2800 3000 3200 3400 3600 3800 4000 4200
400
-1
wavenumber (cm )
UV to the Visible range. So the increase in the amount of amines in the organic materials when the CH4 percentage decreases, could explain the increase of k values measured here. As bands assigned to CH2 and CH3 stretch vibrations increase with the methane concentration, one can deduce that amount of aliphatic methyl increase in tholins when they are produced with higher methane percentage. Based on these results we can affirm that amine-rich polymers are more abundant in tholins produced at lower methane concentration. These results are in agreement with previous work reported for instance by Imanaka et al. (2004). These authors explain the variation of imaginary optical index of tholins by the evolution of the presence of nitrogen rich polymers. In this work, the amount of nitrogen was constant. The C/N ratio in tholins, measured by elemental analysis, shows a dependence of the amount of nitrogen in polymers as function of working total pressure (from 2 to 1200 Pa). In fact at lower pressure, lower C/N ratio was observed than at higher pressure. Using IR and UV–Visible spectroscopies coupled to microprobe laser desorption/ionization mass spectrometry and Raman spectroscopy, Imanaka et al. (2004) suggest also that larger k values observed at lower pressure may also be dueto the presence of the nitrogen-containing polycyclic aromatic compounds in tholins.
5. Comparison with other tholins films produced by a plasma discharge and models based in observational data Sciamma-O’Brien et al. (2012) have reported that a large variation of tholins k values is obtained depending on the experimental methods used. In order to propose a pertinent comparison with other works, we selected only the results of Khare et al. (1984) and Imanaka et al. (2004), who used a low pressure cold plasma discharge with low current density and an initial injected methane concentration of 10%. Fig. 8a presents in linear scale, the k values obtained by Khare et al. (1984) for a pressure of 0.2 mbar and those obtained by Imanaka et al. (2004) for pressures of 0.67 and 1.6 mbar. All k values decrease when wavelength increases. We observe that values obtained by Khare et al. (1984) in the visible range are four times higher than both our values and those reported by Imanaka. In the near Infra Red, all k values are in agreement showing a weak k value of few 103. In the visible range, a common trend is seen in our study and that of Imanaka et al. (2004): an exponential decay followed by an asymptotic behavior with k values of the order of a few 103. We note that k values obtained by Imanaka et al. (2004) at a pressure = 0.6 mbar agree better to our results.
800
1000
wavelength (nm) Fig. 8a. Imaginary part, k, of the refractive index of tholins produced by plasma discharge in a N2/CH4 90/10 mixture: (green) this work at 0.9 mbar, (red) Khare et al. (1984), (black) Imanaka et al. (2004) at 0.67 mbar, (blue) Imanaka et al. (2004) at 1.6 mbar. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
This work (P=0.9 mbar) Khare et al. 1984
1.70
1.65
n
Fig. 7. Absorption spectra of tholins in the Mid-Infrared with different initial CH4 concentrations in the gas mixture.
600
1.60
1.55
1.50 400
500
600
700
800
900
1000
Wavelength (nm) Fig. 8b. Real part, n, of the refractive index of tholins produced by plasma discharge: (green) this work, (red) Khare et al. (1984). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Fig. 8b shows n values obtained by Khare et al. (1984) compared to our values. Unlike Khare et al. (1984) measurements, we do not observe a maximum value of n around 500 nm, but a continuous decay as a function of wavelength and with values lower than obtained by Khare. In Fig. 9, k values obtained for the four samples studied here are compared with values deduced from Titan observations conducted by the Voyager probe (Rages and Pollack, 1980) and Cassini/VIMS instrument (Rannou et al., 2010). The same decay as function of wavelength is observed whether for our values or the values deduced from Titan’s observations. We note that the recent values reported in Rannou et al. (2010) are lower than the previous ones reported by Rages and Pollack (1980), but no explanation is found in Rannou et al. (2010) to explain the difference. We note that observed k values by Rannou et al. (2010) are between those of TF1 and TF2. However this observation is not enough to conclude about the experimental percentage of methane that must be used in the PAMPRE experiments to best mimic the observed optical indices. In fact, the complex index of Titan calculated from Voyager data is uncertain by a factor of about 2 (Sagan
676
A. Mahjoub et al. / Icarus 221 (2012) 670–677
TF2
0.14
k
Table A1 Values of the real and imaginary parts, n and k, of the PAMPRE tholin thin film refractive index from 370 nm to 1000 nm.
TF1
0.16
TF5
0.12
TF10
1%
0.10
Rages et al.1980 Rannou et al.2010
n
k
n
K
n
k
N
k
1.62 1.61 1.61 1.6 1.59 1.59 1.58 1.57 1.57 1.56 1.55 1.54 1.54 1.53 1.53 1.52 1.51 1.51 1.50 1.50 1.50 1.50 1.49 1.49 1.49 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.47 1.47 1.47 1.47 1.47 1.47 1.47 1.47 1.47 1.47 1.47 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48
1.6 E1 1.5E1 1.4E1 1.3E1 1.2E1 1.1E1 1.0E1 9.8E2 9.4E2 8.5E2 8.0E2 7.4E2 7.0E2 6.8E2 6.5E2 5.7E2 5.4E2 5.1E2 4.9E2 4.7E2 4.6E2 4.4E2 4.3E2 4.2E2 4.1E2 4.1E2 4.0E2 3.9E2 3.8E2 3.7E2 3.6E2 3.6E2 3.5E2 3.4E2 3.4E2 3.3E2 3.2E2 3.1E2 3.0E2 3.0E2 3.0E2 2.9E2 2.9E2 2.8E2 2.7E2 2.7E2 2.6E2 2.6E2 2.5E2 2.5E2 2.5E2 2.5E2 2.5E2 2.4E2 2.4E2 2.4E2 2.4E2 2.4E2 2.4E2 2.4E2 2.4E2 23E2 2.3E2 2.3E2
1.64 1.64 1.63 1.63 1.62 1.62 1.61 1.6 1.6 1.59 1.59 1.58 1.57 1.57 1.57 1.56 1.56 1.55 1.55 1.55 1.55 1.54 1.54 1.54 1.54 1.54 1.54 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53
7.3E2 6.8E2 6.1E2 5.5E2 5.0E2 4.5E2 4.0E2 3.6E2 3.3E2 2.9E2 2.6E2 2.4E2 2.2E2 2.0E2 1.8E2 1.7E2 1.6E2 1.5E2 1.4E2 1.3E2 1.2E2 1.2E2 1.2E2 1.1E2 1.1E2 1.1E2 1.0E2 1.0E2 9.8E3 9.7E3 9.5E3 9.3E3 9.1E3 8.6E3 8.4E3 8.1E3 8.0E3 7.8E3 7.6E3 7.6E3 7.5E3 7.3E3 7.1E3 7.0E3 6.7E3 6.7E3 6.6E3 6.5E3 6.5E3 6.2E3 6.2E3 6.0E3 5.9E3 5.9E3 5.8E3 5.7E3 5.7E3 5.7E3 5.7E3 5.4E3 5.4E3 5.4E3 5.4E3 5.4E3
1.67 1.67 1.66 1.65 1.65 1.64 1.65 1.63 1.63 1.62 1.61 1.61 1.61 1.61 1.6 1.6 1.6 1.59 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58
6.0E2 5.4E2 4.9E2 4.2E2 3.8E2 3.4E2 3.0E2 2.8E2 2.4E2 2.2E2 1.9E2 1.8E2 1.6E2 1.4E2 1.3E2 1.1E2 1.0E2 1.0E2 9.3E3 8.8E3 8.1E3 8.0E3 7.6E3 7.4E3 7.0E3 6.9E3 6.7E3 6.5E3 6.3E3 6.2E3 6.1E3 6.0E3 5.8E3 5.7E3 5.6E3 5.5E3 5.4E3 5.3E3 5.3E3 5.2E3 5.0E3 4.9E3 4.8E3 4.8E3 4.7E3 4.6E3 4.6E3 4.5E3 4.4E3 4.4E3 4.3E3 4.2E3 4.2E3 4.0E3 4.0E3 4.0E3 4.0E3 4.0E3 4.0E3 4.0E3 4.0E3 4.0E3 3.9E3 3.9E3
1.67 1.66 1.66 1.66 1.65 1.64 1.64 1.63 1.62 1.62 1.62 1.62 1.61 1.61 1.61 1.60 1.60 1.60 1.59 1.59 1.59 1.59 1.59 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.59 1.59 1.59 1.57 1.57 1.57 1.57 1.57 1.57 1.57 1.57 1.57 1.57
3.6E2 3.3E2 3.0E2 2.7E2 2.4E2 2.1E2 1.9E2 1.7E2 1.6E2 1.5E2 1.4E2 1.2E2 1.1E2 1.0E2 9.3E3 8.6E3 7.8E3 7.2E3 6.6E3 6.2E3 6.0E3 5.7E3 5.4E3 5.2E3 5.1E3 5.0E3 4.9E3 4.7E3 4.6E3 4.5E3 4.4E3 4.3E3 4.2E3 4.1E3 4.0E3 3.9E3 3.9E3 3.8E3 3.7E3 3.6E3 3.6E3 3.5E3 3.5E3 3.5E3 3.4E3 3.3E3 3.2E3 3.1E3 3.1E3 3.0E3 3.0E3 3.0E3 2.9E3 2.9E3 2.8E3 2.8E3 2.8E3 2.8E3 2.8E3 2.8E3 2.8E3 2.8E3 2.8E3 2.7E3
0.08 0.06 0.04 0.02 0.00 300
400
500
600
700
800
900
1000
1100
wavelength (nm) Fig. 9. Tholins films k values compared with Titan’s aerosols deduced from Voyager and Cassini/VIMS observations.
et al., 1992) and imaginary indices obtained for TF2, TF5, TF10 are relatively close to each other. Moreover, we remind that, in continuous working conditions, the CH4 amount into the plasma discharge is half the injected one (Sciamma-O’Brien et al., 2010). The methane consumption is certainly different from a plasma setup to another. The initial methane amount should therefore not be considered as an absolute indicator of the gas mixture in which aerosols are produced. Rannou et al. (2010) values are deduced from DISR data (Tomasko et al., 2008), in low altitudes (<140 m), using a model of scattering by fractal aggregates. k values so obtained are particularly high. These high k values obtained from Titan’s observations are quite in disagreement with aliphatic rich aerosols, whereas a large aliphatic content was suggested by the IR spectra from CIRS and VIMS measurements (Gautier et al., 2012). This apparent disagreement could possibly be explained by a difference in the chemical composition of Titan aerosols in low and high altitudes.
6. Conclusion All our measurements have been done with the same physical experimental conditions (pressure and temperature) while Imanaka et al. (2004) changed the pressure for a given methane percentage. So in this work we characterize the influence of the methane concentration on optical indices of tholins mimicking Titan’s aerosols and produced by a RF plasma discharge. An important decrease of the imaginary optical index, when the CH4 concentration increases, has been demonstrated. Mid-Infrared spectra of thin tholins films produced in different CH4 concentrations allows to connect this k variation to nitrogen-rich polymers preferably produced in lower methane percentages. In Titan’s atmosphere, the CH4 concentration varies with altitude and was varied during the evolution of this atmosphere. In this context, the study presented in this paper is very important to model albedo and then photochemical variations both in space and in time. Nevertheless, we must keep in mind that the initial amount of CH4 injected in the reactor is not the concentration in the continuous working conditions when tholins are produced. In fact, Sciamma et al. (2010) have shown that the amount of CH4 is approximately half of the injected one. Comparison of our results to data deduced from Titan’s observations raises a possible inconsistency between two observations: the mid-IR VIMS and CIRS signatures are in agreement with an aerosol dominated by aliphatic carbon content, whereas the DISR-derived important visible absorption seems to be incompatible with such an important aliphatic content.
370 380 390 400 410 420 430 440 450 460 470 480 490 500 510 520 530 540 550 560 570 580 590 600 610 620 630 640 650 660 670 680 690 700 710 720 730 740 750 760 770 780 790 800 810 820 830 840 850 860 870 880 890 900 910 920 930 940 950 960 970 980 990 1000
2%
5%
10%
Acknowledgments This work was financially supported by CNRS (PNP, ANR09-JCJC-0038 contract). We thank Dr. J.M. Coanga for his cooperation in ellipsometric measurements.
A. Mahjoub et al. / Icarus 221 (2012) 670–677
We thank Dr. P. Pernot for tholins photos and Dr. Dumas for given us the opportunity to use SMIS facility to record Mid-Infrared spectra. The authors are grateful to DGCIS of French Government, Region Ile de France and Conseil Général des Yvelines, CG78 for funding of M-2000V spectroscopic ellipsometer from J.A. Woollam Co. through FUI projects MENKAR and AUDACE from Competitive Cluster MOVEO. Appendix A See Table A1. References Alcouffe, G. et al., 2010. Capacitively Coupled Plasma used to simulate Titan’s atmospheric chemistry. Plasma Sources Sci. Technol. 19, 015008. Atreya, S.K. et al., 2006. Titan’s methane cycle. Planet. Space Sci. 54, 1177–1187. Azzam, R.M.A., Bashara, N.M., 1989. Ellipsometry and Polarized Light. NorthHolland. Berndt, J., Kovacevic, E., Stefanovic, I., Boufendi, L., 2009. J. Appl. Phys. 106, 063309. Carrasco, N. et al., 2009. Chemical characterization of Titan’s Tholins: Solubility, morphology and molecular structure revisited. J. Phys. Chem. A 113, 11195– 11203. Dahoo, P.R. et al., 2004. Evidence by spectroscopic ellipsometry of optical property change in pulsed laser deposited NiO films when heated in air at Neel temperature. Appl. Phys. A: Mater. Sci. Process. 79, 1439–1443. Derenne, S., Coelho, C., Anquetil, C., Szopa, C., Rahman, A.S., McMillan, P.F., Cora, F., Pickard, C.J., Quirico, E., Bonhomme, C., 2012. New insights into the structure and chemistry of Titan tholins via 13C and 15N solid state nuclear magnetic resonance spectroscopy. Icarus 221, 844–853. Dumas, P., Polack, B., Largarde, B., Chubar, O., Giogetta, J.L., Lefrançois, S., 2006. Synchrotron infrared microscopy at the French synchrotron facility SOLEIL. Infrared Phys. Technol. 49, 152–160. Franta, D. et al., 2005. Optical properties of NiO thin films prepared by pulsed laser deposition technique. Appl. Surf. Sci. 244, 420–426. Fujiwara, H., 2007. Spectroscopic Ellipsometry: Principles and Applications. Marusen, Tokyo. Gautier, T., Carrasco, N., Mahjoub, A., Vinatier, S., Giuliani, A., Szopa, C., Anderson, C.M., Correia, J.J., Dumas, P., Cernogora, G., 2012. Mid- and far-infrared absorption spectroscopy of Titan’s aerosols analogues. Icarus 221, 320–327. Hadamcik, E., Renard, J.-B., Alcouffe, G., Cernogora, G., Levasseur-Regourd, A.C., Szopa, C., 2009. Laboratory light scattering measurements with Titan’s aerosols analogues produced by a dusty plasma. Planet. Space Sci. 57, 1631–1641. Hasenkopf, C.A., Beaver, M.R., Trainer, M.G., Dewitt, H.L., Freedman, M.A., Toon, O.B., McKay, C.P., Tolbert, M.A., 2010. Optical properties of Titan and early Earth haze laboratory analogs in the mid-visible. Icarus 207, 903–913. Hébrard, E., Dobrijevic, M., Bélnilin, Y., Raulin, F., 2007. Photochemical kinetic uncertainties in modeling Titan’s atmosphere: First consequences. Planet. Space Sci. 55, 1470–1479. Herzinger, C.M., Johs, B., McGahan, W.A., Woollam, J.A., Paulson, W., 1998. Ellipsometric determination of optical constants for silicon and thermally grown silicon dioxide via a multi-sample, multi-wavelength, multi-angle investigation. J. Appl. Phys. 83, 3323. Horvath, G. et al., 2009. Corona discharge experiments in admixtures of N2 and CH4: A laboratory simulation of Titan’s atmosphere. Plasma Sources Sci. Technol. 18, 034016.
677
Horvath, G., Mason, N.J., Polachova, L., Zahoran, M., Moravsky, L., Matejcik, S., 2010. Packed bed DBD discharge experiments in admixtures of N2 and CH4. Plasma Chem. Plasma Process. 30, 565–577. Imanaka, H., Khare, B.N., Elsila, J.E., Bakes, E.L.O., McKay, C.P., Cruikshank, D.P., Sugita, S., Matsui, T., Zare, R.N., 2004. Laboratory experiments of Titan tholin formed in cold plasma at various pressures: Implications for nitrogencontaining polycyclic aromatic compounds in Titan haze. Icarus 168, 344–366. Jellison, G.E., Modine Jr., F.A., 1996. Parameterization of the optical functions of amorphous materials in the interband region. Appl. Phys. Lett. 69, 371–373. Khare, B.N., Sagan, C., Arakawa, E.T., Suits, F., Callcott, T.A., Williams, M.W., 1984. Optical constants of organic tholins produced in a simulated titanian atmosphere: From X-ray to microwave frequencies. Icarus 60, 127–137. Kramers, H.A., 1927. La diffusion de la lumiere par les atomes, AttiCong. Intern. Fisica, (Transactions of Volta centenary congress). Como 2, 545–557. Kronig, R.de L., 1926. On the theory of the dispersion of X-rays. J. Opt. Soc. Am. 12, 547–557. Loutete-Dangui, E.D., Varret, F., Codjovi, E., Dahoo, P.R., Boukheddaden, K., 2007. Thermal spin transition in Fe[NH3]Br2 investigated by spectroscopic ellipsometry. Phys. Rev. B 75, 184425-1–184425-7. McKay, C.P., Pollack, J.B., Courtin, R., 1989. The thermal structure of Titan’s atmosphere. Icarus 80, 23–53. Pernot, P., Carrasco, N., Thissen, R., Schmitz-Afonso, I., 2010. Tholinomics-chemical analysis of nitrogen-rich polymers. Anal. Chem. 82, 1371–1380. Quirico, E., Montagnac, G., Lees, V., McMillan, P.F., Szopa, C., Cernogora, G., Rouzaud, J.-N., Simon, P., Bernard, J.-M., Coll, P., 2008. New experimental constraints on the composition and structure of tholins. Icarus 198, 218–231. Rages, K., Pollack, J.B., 1980. Titan aerosols: Optical proprieties and vertical distribution. Icarus 41, 119–130. Ramirez, S.I., Coll, P., da Silva, A., Navarro-González, R., Lafait, J., Raulin, F., 2002. Complex refractive index of Titan’s aerosol analogues in the 200–900 nm domain. Icarus 156, 515–529. Rannou, P., Cours, T., Le Mouélic, S., Rodriguez, S., Sotin, C., Drossart, P., Brown, R., 2010. Titan haze distribution and optical properties retrieved from recent observations. Icarus 208, 850–867. Sagan, C., Thompson, W.R., Khare, B.N., 1992. Titan: A laboratory for prebiological organic chemistry. Acc. Chem. Res. 25, 286–292. Samuelson, R.E., 1983. Radiative equilibrium model of Titan’s. Icarus 53, 364–387. Sciamma-O’Brien, E., Carrasco, N., Szopa, C., Buch, A., Cernogora, G., 2010. Titan’s atmosphere: An optimal gas mixture for aerosol formation? Icarus 209, 704– 714. Sciamma-O’Brien, E., Dahoo, P.R., Hadamcik, E., Carrasco, N., Quirico, E., Szopa, C., Cernogora, G., 2012. Optical constants from 370 nm to 900 nm of Titan tholins produced in a low pressure RF plasma discharge. Icarus 218, 356–363. Szopa, C., Cernogora, G., Boufendi, L., Correia, J.-J., Coll, P., 2006. PAMPRE: A dusty plasma experiment for Titan’s tholins production and study. Planet. Space Sci. 54, 394–404. Tobie, G., Lunine, J., Sotin, C., 2006. Episodic outgassing as the origin of atmospheric methane on Titan. Nature 440 (7080), 61–64. Tomasko, M. et al., 2008. A model of Titan’s aerosols based on measurements made inside the atmosphere. Planet. Space Sci. 56, 669–707. Tran, B.N., Joseph, J.C., Ferris, J.P., Persan, P.D., Chera, J.J., 2003. Simulation of haze formation using a photochemical flow reactor. The optical constants of the polymers. Icarus 165, 379–390. Vuitton, V., Tran, B.N., Persans, P.D., Ferris, J.P., 2009. Determination of the complex refractive indices of Titan haze analogs using photothermal deflection spectroscopy. Icarus 203, 663–671. Waite, J.H. et al., 2007. The process of tholins formation in Titan’s upper atmosphere. Science 316, 870–875. Weber, J.W., Hansen, T.A.R., van de Sanden, M.C.M., Engeln, R., 2009. B-Spline parametrisation of the dielectric function applied to spectroscopic ellipsometry on amorphous carbon. J. Appl. Phys. 106, 3460–3478.
Contents lists available at SciVerse ScienceDirect
Icarus journal homepage: www.elsevier.com/locate/icarus
Influence of methane concentration on the optical indices of Titan’s aerosols analogues A. Mahjoub ⇑, N. Carrasco, P.-R. Dahoo, T. Gautier, C. Szopa, G. Cernogora Université Versailles St-Quentin, UPMC Univ. Paris 06, CNRS, LATMOS, 11 Blvd. d’Alembert, 78280 Guyancourt, France
a r t i c l e
i n f o
Article history: Received 17 February 2012 Revised 16 August 2012 Accepted 16 August 2012 Available online 11 September 2012 Keywords: Titan Photometry Radiative transfer Titan, atmosphere
a b s t r a c t This work deals with the optical constant characterization of Titan aerosol analogues or ‘‘tholins’’ produced with the PAMPRE experimental setup and deposited as thin films onto a silicon substrate. Tholins were produced in different N2–CH4 gaseous mixtures to study the effect of the initial methane concentration on their optical constants. The real (n) and imaginary (k) parts of the complex refractive index were determined using the spectroscopic ellipsometry technique in the 370–1000 nm wavelength range. We found that optical constants depend strongly on the methane concentrations of the gas phase in which tholins are produced: imaginary optical index (k) decreases with initial CH4 concentration from 2.3 102 down to 2.7 103 at 1000 nm wavelength, while the real optical index (n) increases from 1.48 up to 1.58 at 1000 nm wavelength. The larger absorption in the visible range of tholins produced at lower methane percentage is explained by an increase of the secondary and primary amines signature in the mid-IR absorption. Comparison with results of other tholins and data from Titan observations are presented. We found an agreement between our values obtained with 10% methane concentration, and Imanaka et al. (Imanaka, H., Khare, B.N., Elsila, J.E., Bakes, E.L.O., McKay, C.P., Cruikshank, D.P., Sugita, S., Matsui, T., Zare, R.N. [2004]. Icarus, 168, 344–366) values, in spite of the difference in the analytical method. This confirms a reliability of the optical properties of tholins prepared with various setups but with similar plasma conditions. Our comparison with Titan’s observations also raises a possible inconsistency between the mid-IR aerosol signature by VIMS and CIRS Cassini instruments and the visible Huygens-DISR derived data. The mid-IR VIMS and CIRS signatures are in agreement with an aerosol dominated by an aliphatic carbon content, whereas the important visible absorption derived from the DISR measurement seems to be incompatible with such an important aliphatic content, but more compatible with an amine-rich aerosol. Ó 2012 Elsevier Inc. All rights reserved.
1. Introduction Titan, the biggest Saturn’s moon, has an important astrobiological interest because of its dense atmosphere unique among all Solar System satellites. Titan’s atmosphere is mainly constituted of nitrogen and methane. The energetic particles coming from Saturn’s magnetosphere and the solar UV radiation gives enough energy to dissociate both nitrogen and methane molecules, allowing a complex chemistry to take place (Waite et al., 2007). This complex photochemistry leads to the formation of solid aerosols as a permanent haze around the satellite. Titan’s aerosols play an important role in the climate, the composition and the properties of Titan’s atmosphere as well as its surface. This haze plays an important role in the radiative transfer in the atmosphere of Titan (Samuelson, 1983) and strongly influences its thermal structure (McKay et al., 1989; Rannou et al., 2010). Their optical and physical ⇑ Corresponding author. E-mail address: [email protected] (A. Mahjoub). 0019-1035/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.icarus.2012.08.015
properties are essential to analyze the observational data acquired by the Cassini Orbiter and Huygens probe instruments, among others. Models developed to explain the optical properties of Titan’s atmosphere require knowledge of optical indices of Titan’s haze (see for instance Tomasko et al., 2008 or Rannou et al., 2010). For these reasons, determining the optical properties of the haze is critical in the analysis of the observational data. As no return sample is planned from Titan’s atmosphere in the short term, the use of laboratory experiments is of primary importance to produce analogues of Titan’s aerosols. Indeed, the study of these analogues and of their production mechanisms enables one to provide important information on the processes taking place in Titan’s atmosphere. Several experimental devices have been used as photochemical reactors or plasma discharges. The latter are the most efficient processes to produce solid material named tholins. Several discharges have been used up to now: Direct Current discharges (DC) (Khare et al., 1984), barrier discharges (Horvath et al., 2010), corona discharges (Horvath et al., 2009), Radio frequency discharges
A. Mahjoub et al. / Icarus 221 (2012) 670–677
Inductively coupled (Imanaka et al., 2004) or capacitively coupled as what was used in the present work on the PAMPRE reactor (Szopa et al., 2006). In the PAMPRE reactor tholins are produced as powder (Szopa et al., 2006) or as thin films on substrates (Quirico et al., 2008; Sciamma-O’Brien et al., 2012). Some properties of produced tholins have been studied: particles size and morphology (Hadamcik et al., 2009), production efficiency and elemental analysis (Sciamma-O’Brien et al., 2010), infrared spectroscopy (Quirico et al., 2008; Gautier et al., 2012), solubility (Carrasco et al., 2009), solid NMR (Derenne et al., 2012) and mass spectrometry (Pernot et al., 2010; Carrasco et al., 2009). In Titan’s upper atmosphere (ionosphere) where aerosols are also produced, the CH4 varies from 2% to 10% (Hébrard et al., 2007). Thus, it is of utmost importance to determine the optical constants of tholins synthesized from N2–CH4 gas mixtures with different CH4 concentrations. Moreover, Atreya et al. (2006) calculate that without methane emission source, photochemistry would lead to the irreversible loss of methane from Titan’s atmosphere within 10–100 myr range. A methane emission source is therefore proposed by Tobie et al. (2006) as episodic outgassing processes. Such an irregular source involves a varying atmospheric methane concentration during the evolution of Titan’s atmosphere. The atmospheric aerosols would therefore have been produced in different methane conditions through time. We suspect that the aerosol optical indices are sensitive to this parameter and could imply a change of the optical opacity of the atmosphere. Studying the influence of a varying methane percentage on the aerosol optical indices is thus crucial to understand the evolutionary history of Titan and its atmosphere. Many of plasma discharges producing tholins have been working with an initial 10:90 CH4:N2 mixture (see for instance Khare et al., 1984, or Imanaka et al., 2004). Other works have been made with other initial methane amounts (Khare et al., 1984; Imanaka et al., 2004; Tran et al., 2003; Ramirez et al., 2002; Vuitton et al., 2009). Nevertheless, these experiments have been performed with different experimental setups and the optical indices have often been measured with different analytical techniques. To our knowledge, there is no consistent study of the evolution of the optical properties with the methane concentration. In the present work, tholins are produced in the PAMPRE plasma discharge device. Sciamma-O’Brien et al. (2010) have shown that a stationary state (continuous working conditions) is reached within about 2 min after the beginning of the plasma discharge. In the stationary state the amount of CH4 is approximately half of the injected one, during the film growth (for a RF power of 30 W, 55 sccm gas flow at a pressure of 0.9 mbar). It was also shown that the elemental composition of tholins (powder) changes with the increase of the initial amount of CH4 injected: C content remains constant at a value about 30% whereas N amount decreases from 30% down to 15% and the H amount increases from 30% up to 45%. So a change of the optical constants as a function of initial CH4 concentration is expected. Optical properties of tholins films have previously been studied for the injected gas mixture CH4:N2 = 5:95 (Sciamma-O’Brien et al., 2012). The question of the effect of tholins production conditions on their optical properties remains open. Many studies have been carried out to determine the real and imaginary parts of the refractive index of tholins produced by various experimental techniques (photochemistry, electric discharges). These indices have been determined using different methods: ellipsometry, spectrophotometric measurements (Khare et al., 1984; Imanaka et al., 2004; Tran et al., 2003; Ramirez et al., 2002; Vuitton et al., 2009; Sciamma-O’Brien et al., 2012), and recently cavity ring down aerosol extinction spectroscopy (Hasenkopf et al., 2010). Sciamma-O’Brien et al. (2012) shows that the optical constants obtained with such different techniques are
671
quite different. In fact, the wide variety of initial gas mixtures used in these experiments produced tholins with different elemental composition which are therefore expected to have different optical properties. In addition, different techniques of refractive index measurement and different experimental conditions (e.g. pressure or temperature) can influence the resulting tholins optical properties. In this paper we determine the imaginary and real parts of the complex optical index of tholins produced with four different CH4 concentrations under the same experimental conditions. Tholins are deposited as thin films onto the same substrate: silicon (Si). Optical properties of the four samples are determined using ellipsometry technique. This study allows the investigation of the influence of the methane concentration independently of substrate and optical diagnostic technique. 2. Experimental methodology 2.1. Tholins film production PAMPRE is an experimental cold plasma discharge described in details in previous publications (Szopa et al., 2006; Alcouffe et al., 2010). Briefly a Capacitively Coupled Plasma radio frequency (CCP RF) discharge at 13.56 MHz is driven between a polarized electrode and a grounded cylindrical grid box confining the plasma. The injected CH4:N2 gas mixture can be changed from 0:100 to 10:90. PAMPRE experimental setup allows the production of tholins in volume (tholins grow in suspension in the gas phase) as well as deposited as thin films onto solid substrates. The continuous gas flow rate gives a constant pressure of 0.9 mbar in the reactor during the tholin production. The reactive gas mixture is at room temperature. In the present work, we focused on the study of optical properties of tholins films deposited on substrates as a function of the initial gas mixture injected in the plasma reactor. For this purpose a thin tholins film was deposited onto a 0.5 mm thick and 12 mm diameter polished silicon substrate. We have chosen silicon substrates because silicon is often used for microelectronic devices and its optical properties are well established. The substrate was placed at the center of the grounded electrode inside the plasma confining grid box. The experiment was run for 2 h in order to produce a film no more than 1 lm thick and therefore limit the optical absorption. After tholins film deposition, samples are studied by spectroscopic ellipsometry at atmospheric pressure in order to determine the real and imaginary parts of their refractive index. 2.2. Spectroscopic ellipsometry Spectroscopic ellipsometry is an optical technique based on the measurement of the change in polarized light upon light reflection on a sample (Azzam and Bashara, 1989; Fujiwara, 2007). Because of recent advances in computer technology, the spectroscopic ellipsometry technique has developed rapidly. As a consequence, ellipsometry is actually used routinely to characterize thin films in electronic and optical industry. In spectroscopic ellipsometry, linearly polarized light is incident at an oblique angle onto the sample. The reflected light from the sample surface is, in general, elliptically polarized. The measured quantity in ellipsometry is the so called complex reflectance ratio, expressed as
q¼
rp ¼ tanðwÞeiD rs
ð1Þ
where rp and rs are the complex Fresnel reflection coefficients for p- and s-polarized light, respectively; tan(w) is the ratio of the
672
A. Mahjoub et al. / Icarus 221 (2012) 670–677
amplitude changes and D is the relative difference in phase changes for the p- and s-polarized light components upon reflection on the sample surface. From experimentally determined q, information can be obtained about the optical properties of the material under study in terms of number of layers, their thicknesses, their roughness and the optical constants of each layer at each wavelength. Spectroscopic ellipsometry is an inverse problem, where parameters w and D are measured but the sample optical properties must be obtained by a model (Dahoo et al., 2004; Franta et al., 2005; Loutete-Dangui et al., 2007). Model-generated ellipsometric parameters (w and D) can then be compared to experimental data. The sample optical properties (fit parameters) whose response best matches the experimental data are found using a regression analysis. The mean squared error (MSE) and comparison between the model-generated curves and experimental data are used to estimate the fit quality of the final model. In the study presented here, we used an M-2000V spectroscopic ellipsometer from J.A. Woollam Co. The M-2000V is a rotating compensator ellipsometer with CCD camera to measure all wavelengths simultaneously. Data are collected 20 times per second. To optimize the signal-to-noise ratio, we chose a 100 s long measurement duration for each full spectrum. The light source is a tungsten lamp which covers a wide spectral range from 370 nm to 1000 nm. Within this spectral range, spectroscopic ellipsometry data are collected over 400 points. We used an angle of incidence of 70°. At that angle, the light spot reflected from the tholins film was approximately 8 mm 3 mm, i.e. smaller than the substrate diameter. The spectroscopic ellipsometry measurements were then analyzed using Woollam’s Complete-EASE™ analysis software (Complete-EASE™ Data Analysis Manual By J.A. Woollam Co Inc. June 15, 2008). With this software, it is possible to describe the different layers composing a film using a specific model for each layer. In the Complete-EASE™ software, morphologic characteristics are also considered in the fitting calculation: surface roughness, thickness non-uniformity as well as the angle offset added to the nominal angle of incidence to compensate for small angle alignment errors. The surface roughness is determined by optimizing a mixed external layer composed of air and tholins. The thickness non-uniformity is related to the polarization variation on the whole ellipsometric spot.
For the Si substrate optical indices are well known (see for instance Herzinger et al., 1998), and tabulated data available in the CompleteEASE™ software have been used. As optical properties of tholins are unknown, a parameteric model must be employed. Hadamcik et al. (2009) showed that tholins produced with PAMPRE setup have an amorphous structure. Then, in order to describe tholins film, we have used a Tauc–Lorentz model which is well adapted for absorbing amorphous organic films (Jellison and Modine, 1996). In the Tauc–Lorentz model the imaginary part of the complex relative dielectric constant e = e1 + ie2 is given by:
e2 ¼
AE0 CðE Eg Þ2 2
ðE
E20 Þ2
1 ; for E > Eg þC E E
ð2Þ
2 2
e2 ¼ 0; for E < Eg This expression has four energy parameters: the band gap Eg, the peak in the joint density of states E0, the broadening parameter C, and the prefactor A, which includes the optical transition matrix elements (Jellison and Modine, 1996). The real part of the dielectric function is derived from e2 using the Kramers–Kronig relations (Kramers, 1927; Kronig, 1926):
2
e1 ðxÞ ¼ 1 þ P p
Z
1
0
x0 e2 ðx0 Þ 0 dx x02 x2
ð3Þ
where P is the Cauchy principal value of the integral:
P
Z 0
1
Z f ðx0 Þdx0 ¼ lim d!0
0
xd
f ðx0 Þdx0 þ
Z
1
f ðx0 Þdx0
ð4Þ
xþd
The complex refractive indices are then deduced from the relative dielectric constants by the relation:
e ¼ e1 þ ie2 ¼ ðn þ ikÞ2
ð5Þ
In a previous work, Sciamma-O’Brien et al. (2012) studied similar tholins films produced in PAMPRE setup with a 5:95 CH4:N2 gas mixture deposited onto a Al:SiO2 substrate. In this case, the organic layer was modeled using a mathematical fitting named B-Spline model. The question of the influence of the chosen model on the results has therefore to be addressed. 3.2. Comparison between B-Spline and Tauc–Lorentz fitting models
3. Ellipsometry data analysis 3.1. Data analyzing procedure Samples studied in this work are thin films deposited on Si substrates. The measured ellipsometric parameters w and D result from the interaction of the light with both tholins film and Si substrate. Moreover the interface between film and the ambient air presents some roughness. Then the sample must be treated with a multilayer model, where the interface is considered as an additional layer composed by a mixture of organic material and air. In this work, a model consisting of three layers has been defined (Fig. 1).
Fig. 1. Schematic representation of a layered optical model with tholins film.
Both models, B-Spline and Tauc–Lorentz, can be used to characterize optical properties of absorbing materials (Weber et al., 2009). In order to evaluate the influence of the chosen model on the optical results we have used the w and D parameters recorded by Sciamma-O’Brien et al. (2012) and calculated both thickness and complex index of the tholins films using a Tauc–Lorentz model. With the Tauc–Lorentz model, we obtain a film thickness of 420 ± 15 nm in agreement with the B-Spline model (Sciamma-O’Brien et al., 2012). Fig. 2 shows in linear scales both real and imaginary parts of the optical index obtained by Tauc–Lorentz and B-Spline models. The values obtained with both models are in good agreement in the 370–900 nm range. Sciamma-O’Brien et al. (2012) point-out a validity limit of the B-Spline model, which prevents the k determination in the quasi-transparent region above 900 nm. In fact, as Tauc–Lorentz model is based on a physical description of the dielectric constants, no numerical divergence appears above 900 nm. This makes possible the determination of optical indices in the range 900–1000 nm (see Fig. 2). Moreover the Tauc–Lorentz model can give useful information about the electronic transition of organic materials constituting tholins. So, for the following, the Tauc–Lorentz model has been chosen for tholins films description.
673
A. Mahjoub et al. / Icarus 221 (2012) 670–677
0.018
1.8
Tauc-Lorentz model Bspline model
Tauc-Lorentz model Bspline model
1.7
n
k
0.009
0.000
1.6 1.5
400
600
800
1000
1.4
Wavelength (nm)
600
900
Wavelength (nm)
Fig. 2. Comparison between imaginary (left) and real (right) index obtained by B-Spline and Tauc–Lorentz fitting models for the tholins film layer on AlSiO2 produced in Sciamma-O’Brien et al. (2012).
Fig. 3. Comparison between experimental and model generated w and D spectroscopic ellipsometry parameters recorded for TF5.
3.3. Effect of methane concentration on tholins optical properties Ellipsometric measurements recorded for samples produced in different methane concentrations (1%, 2%, 5% and 10%) are analyzed using the model described in Section 3.1. We subsequently used the following convention to refer to the different tholins films: TFx, x corresponding to the CH4 percentage. As an example, we will discuss hereafter the fitting quality obtained for the TF5 sample. Fig. 3 shows the comparison between experimental D and w parameters and those generated by the model described in Section 3.1. Although many parameters are taken into account by the Complete-EASE™ software (model for each layer, non-uniformity, roughness, etc.), a global fitting quality is quantified by the mean square error (MSE). The agreement between the fits and the experimental measurements is validated by a MSE value of 4 103 to 1.4 102. In our case the MSE value calculated by the Complete-EASE™ software is about 5 103 confirming a good agreement between the fits and the experimental measurements. Table 1 summarizes the results of geometrical properties of the films, the accuracy of the fit and the gap energy. The uncertainty of
the film thickness is given by the non-uniformity calculated by Complete-EASE™ software. On Table 1 are also reported the uncertainties on all the Tauc– Lorentz parameters evaluated by the Complete-EASE™ software. Table 1 shows that the major contributor to the error is by far the uncertainty in the film roughness (between 5% and 9%). We note that the uncertainty on all fitting parameters is less than 9%, this indicates that the model used to determine n and k values is a good representation of studied samples. We note that the thickness of the thin tholins films increases with the injected amount of CH4. For powder, Sciamma-O’Brien et al. (2010) have found a maximum of mass production for 4% of injected CH4 for the same pressure condition as the present work. Therefore the film thickness does not follow the same evolution with the methane concentration as the powder mass production into the gas volume. This difference is not explained yet, but is beyond the present study. If we compare the film thickness (420 ± 15 nm) obtained by Sciamma-O’Brien et al. (2012) for deposition onto commercial mirrors and those of present films deposited on Si substrate (850 ± 30 nm) for the same experimental conditions and deposition duration, we observe than the deposition on Si is more efficient by a factor of 2. This effect is probably due to the electric properties of the substrates and the polarization on the electrodes. In the PAMPRE CCP discharge, the driven electrode reaches a negative self-bias potential (Alcouffe et al., 2010) when the electrode where the substrates are displayed is grounded. Then, the grounded electrode receives a flux of negative species: electrons and negative ions. Mirrors used by Sciamma-O’Brien et al. (2012) are isolating material, when Si substrates used in the present work are semi conductive. As a consequence the flux of negative ions on Si is more intense. As negative ions are important in dusty plasma chemistry (Berndt et al., 2009) this could explain why tholins films are thicker on Si than on mirrors. Fig. 4 shows the n values obtained for different TFx samples as function of wavelength. We note that when CH4 concentration
Table 1 Results given by the model for different TFx (corresponding to the CH4 percentage) samples: geometrical properties of the films, the accuracy of the fit and gap energy of tholins films. Sample
Thickness (nm)
Roughness (nm)
MSE
Eg (eV)
A (eV)
E0
C
TF1 TF2 TF5 TF10
628 ± 15 770 ± 25 850 ± 30 910 ± 40
24 ± 2.0 34 ± 2.2 15 ± 0.8 10 ± 0.9
4.5 103 5.2 103 6.2 103 6 103
2.05 ± 0.15 2.11 ± 0.13 2.13 ± 0.15 2.26 ± 0.14
40.1 ± 0.84 40.7 ± 1.04 40.5 ± 0.93 39.73 ± 0.98
12.4 ± 0.12 13.37 ± 0.13 13.9 ± 0.13 15 ± 0.15
24.4 ± 0.5 24.4 ± 0.5 24.3 ± 0.5 24.3 ± 0.5
674
A. Mahjoub et al. / Icarus 221 (2012) 670–677
1.70 1.68
TF10
1.66
TF5
1.64
TF2
1.62
TF1
bright yellow for 10% of CH4 (Fig. 6). Otherwise, we note an increase of the optical band gap Eg (Table 1) for increasing methane concentration. The quantity Eg correlates with the fraction of sp3bonded carbon atoms, with higher values of Eg corresponding to higher values of the sp3 fraction (see for instance Jellison et al. (1996)). This result is an agreement with IR measurements presented in Section 4. The averaged n and k values from 370 nm to 900 nm by 10 nm increments are given in the Appendix A. We note that any quantitative determination of uncertainty is reported in Appendix A.
n
1.60 1.58 1.56 1.54 1.52 1.50
4. About the relationship between CH4 concentration and the chemical composition of thin films
1.48 1.46 400
500
600
700
800
900
1000
Wavelength (nm) Fig. 4. Real part, n, of the tholins thin films refractive indices produced with different initial methane concentration.
TF1 TF2 TF5
0,1
k
TF10
0,01
400
500
600
700
800
900
1000
Wavelength Fig. 5. Imaginary part, k, of the tholins thin films refractive indices produced with different initial methane concentration.
Results obtained in this paper in terms of n and k values in different CH4 percentages are not self-sufficient to correlate the tholins composition to the optical indices without any information about the chemical composition of these organic materials. To strengthen our discussion on the relationship between the composition of tholins and its optical constants, Mid-Infrared spectra were recorded for tholins produced for the different studied methane concentrations. This Infra Red spectroscopic study is very helpful to describe the evolution of functional groups existing in tholins in order to connect it to absorption in the UV–Visible wavelengths region. Spectra were recorded at the SMIS (Spectroscopy and Microscopy in the Infra Red using Synchrotron) beam line of SOLEIL synchrotron radiation facility, France (Dumas et al., 2006). We used a NicPlan microscope coupled to a Nicolet Magna System 560 Fourier Transformed Infra Red (FTIR) spectrometer with a spectral resolution of 4 cm1. The Infra Red sources used for the present work were the synchrotron line for Mid-Infrared. The detector used was the Mercury–Cadmium–Telluride (MCT) detector of the microscope (Mid IR). Analysis were performed in transmission mode on thin tholins films deposited onto MirrIR substrates (Low-e microscope slides from Kevley Technologies) substrates. For each sample three spectra were collected and blanks were performed on a blank substrate. Using this experimental protocol we have measured the transmitted Infra Red intensity It given by:
It ¼ I0 e2ad
CH 4 % Fig. 6. Colors of powder tholins produced in different CH4 concentrations. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
increases, the real optical index increases from 1.48 for TF1 to 1.58 for TF10 at 1000 nm. The real indices so obtained have reasonable values compared to other measurements of optical indices of tholins films (Sciamma-O’Brien et al., 2012). Fig. 5 shows the values of the imaginary parts of the optical indices obtained for tholins films produced at different percentages of injected methane. On these spectra we observe that k values decrease when the concentration of CH4 increases in agreement with the anti-correlated evolution of the n value: for example at 1000 nm the difference in k value is one order of magnitude between TF1 and TF10 (2.3 102 versus 2.7 103). These results are correlated to the colors of tholins produced in different CH4 concentration: the color varies from dark brown at 1% of CH4 to
ð6Þ
where I0 is the incident beam intensity, a is the absorption coefficient and d is the film thickness. The factor 2 in the equation is due to the transmission measurement method: the beam crosses the tholins film two times. The film thickness d is determined by spectroscopic ellipsometry on equivalent substrates (for more details see Gautier et al., 2012). The values are respectively 440 nm; 490 nm; 550 nm and 580 nm, for CH4 amounts of 1%; 2%; 5% and 10%. Fig. 7 shows the absolute comparison of the linear absorption coefficient a in the 2500–4000 cm1 range as a function of the initial methane concentration. In all these spectra we observe a broad doublet centered at 3205 and 3325 cm1 respectively. These bands are attributed to primary and secondary amines. A system of three bands appears at 2871, 2930 and 2970 cm1 these bands being attributed to CH2 and CH3 stretching modes. Comparison of spectra revealed a dramatic dependence of the intensities of different bands as function of the injected methane percentage. Intensities of NH and NH2 bands decrease as methane concentration increases on one hand and intensities of CH2 and CH3 bands increase with methane concentration on the other. These observations are very helpful to understand the dependence of tholins absorption in the UV–Visible range. Indeed, the very broad absorption centered around 220 nm due to the n ? r⁄ transition of amines groups is known to be relatively intense from the
675
A. Mahjoub et al. / Icarus 221 (2012) 670–677
0.10
NH,NH2
3000
1% 2% 5% 10 %
2500 CH2, CH3
0.08
0.06
1500
K
-1
α (cm )
2000
this work Khare et al. (1984) (P=0.2 mbar) Imanaka et al. (2004) (P=1.6 mbar) Imanaka et al. (2004)(P= 0.64 mbar)
0.04
1000 500
0.02
0 0.00
-500 2400 2600 2800 3000 3200 3400 3600 3800 4000 4200
400
-1
wavenumber (cm )
UV to the Visible range. So the increase in the amount of amines in the organic materials when the CH4 percentage decreases, could explain the increase of k values measured here. As bands assigned to CH2 and CH3 stretch vibrations increase with the methane concentration, one can deduce that amount of aliphatic methyl increase in tholins when they are produced with higher methane percentage. Based on these results we can affirm that amine-rich polymers are more abundant in tholins produced at lower methane concentration. These results are in agreement with previous work reported for instance by Imanaka et al. (2004). These authors explain the variation of imaginary optical index of tholins by the evolution of the presence of nitrogen rich polymers. In this work, the amount of nitrogen was constant. The C/N ratio in tholins, measured by elemental analysis, shows a dependence of the amount of nitrogen in polymers as function of working total pressure (from 2 to 1200 Pa). In fact at lower pressure, lower C/N ratio was observed than at higher pressure. Using IR and UV–Visible spectroscopies coupled to microprobe laser desorption/ionization mass spectrometry and Raman spectroscopy, Imanaka et al. (2004) suggest also that larger k values observed at lower pressure may also be dueto the presence of the nitrogen-containing polycyclic aromatic compounds in tholins.
5. Comparison with other tholins films produced by a plasma discharge and models based in observational data Sciamma-O’Brien et al. (2012) have reported that a large variation of tholins k values is obtained depending on the experimental methods used. In order to propose a pertinent comparison with other works, we selected only the results of Khare et al. (1984) and Imanaka et al. (2004), who used a low pressure cold plasma discharge with low current density and an initial injected methane concentration of 10%. Fig. 8a presents in linear scale, the k values obtained by Khare et al. (1984) for a pressure of 0.2 mbar and those obtained by Imanaka et al. (2004) for pressures of 0.67 and 1.6 mbar. All k values decrease when wavelength increases. We observe that values obtained by Khare et al. (1984) in the visible range are four times higher than both our values and those reported by Imanaka. In the near Infra Red, all k values are in agreement showing a weak k value of few 103. In the visible range, a common trend is seen in our study and that of Imanaka et al. (2004): an exponential decay followed by an asymptotic behavior with k values of the order of a few 103. We note that k values obtained by Imanaka et al. (2004) at a pressure = 0.6 mbar agree better to our results.
800
1000
wavelength (nm) Fig. 8a. Imaginary part, k, of the refractive index of tholins produced by plasma discharge in a N2/CH4 90/10 mixture: (green) this work at 0.9 mbar, (red) Khare et al. (1984), (black) Imanaka et al. (2004) at 0.67 mbar, (blue) Imanaka et al. (2004) at 1.6 mbar. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
This work (P=0.9 mbar) Khare et al. 1984
1.70
1.65
n
Fig. 7. Absorption spectra of tholins in the Mid-Infrared with different initial CH4 concentrations in the gas mixture.
600
1.60
1.55
1.50 400
500
600
700
800
900
1000
Wavelength (nm) Fig. 8b. Real part, n, of the refractive index of tholins produced by plasma discharge: (green) this work, (red) Khare et al. (1984). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Fig. 8b shows n values obtained by Khare et al. (1984) compared to our values. Unlike Khare et al. (1984) measurements, we do not observe a maximum value of n around 500 nm, but a continuous decay as a function of wavelength and with values lower than obtained by Khare. In Fig. 9, k values obtained for the four samples studied here are compared with values deduced from Titan observations conducted by the Voyager probe (Rages and Pollack, 1980) and Cassini/VIMS instrument (Rannou et al., 2010). The same decay as function of wavelength is observed whether for our values or the values deduced from Titan’s observations. We note that the recent values reported in Rannou et al. (2010) are lower than the previous ones reported by Rages and Pollack (1980), but no explanation is found in Rannou et al. (2010) to explain the difference. We note that observed k values by Rannou et al. (2010) are between those of TF1 and TF2. However this observation is not enough to conclude about the experimental percentage of methane that must be used in the PAMPRE experiments to best mimic the observed optical indices. In fact, the complex index of Titan calculated from Voyager data is uncertain by a factor of about 2 (Sagan
676
A. Mahjoub et al. / Icarus 221 (2012) 670–677
TF2
0.14
k
Table A1 Values of the real and imaginary parts, n and k, of the PAMPRE tholin thin film refractive index from 370 nm to 1000 nm.
TF1
0.16
TF5
0.12
TF10
1%
0.10
Rages et al.1980 Rannou et al.2010
n
k
n
K
n
k
N
k
1.62 1.61 1.61 1.6 1.59 1.59 1.58 1.57 1.57 1.56 1.55 1.54 1.54 1.53 1.53 1.52 1.51 1.51 1.50 1.50 1.50 1.50 1.49 1.49 1.49 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.47 1.47 1.47 1.47 1.47 1.47 1.47 1.47 1.47 1.47 1.47 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48 1.48
1.6 E1 1.5E1 1.4E1 1.3E1 1.2E1 1.1E1 1.0E1 9.8E2 9.4E2 8.5E2 8.0E2 7.4E2 7.0E2 6.8E2 6.5E2 5.7E2 5.4E2 5.1E2 4.9E2 4.7E2 4.6E2 4.4E2 4.3E2 4.2E2 4.1E2 4.1E2 4.0E2 3.9E2 3.8E2 3.7E2 3.6E2 3.6E2 3.5E2 3.4E2 3.4E2 3.3E2 3.2E2 3.1E2 3.0E2 3.0E2 3.0E2 2.9E2 2.9E2 2.8E2 2.7E2 2.7E2 2.6E2 2.6E2 2.5E2 2.5E2 2.5E2 2.5E2 2.5E2 2.4E2 2.4E2 2.4E2 2.4E2 2.4E2 2.4E2 2.4E2 2.4E2 23E2 2.3E2 2.3E2
1.64 1.64 1.63 1.63 1.62 1.62 1.61 1.6 1.6 1.59 1.59 1.58 1.57 1.57 1.57 1.56 1.56 1.55 1.55 1.55 1.55 1.54 1.54 1.54 1.54 1.54 1.54 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53 1.53
7.3E2 6.8E2 6.1E2 5.5E2 5.0E2 4.5E2 4.0E2 3.6E2 3.3E2 2.9E2 2.6E2 2.4E2 2.2E2 2.0E2 1.8E2 1.7E2 1.6E2 1.5E2 1.4E2 1.3E2 1.2E2 1.2E2 1.2E2 1.1E2 1.1E2 1.1E2 1.0E2 1.0E2 9.8E3 9.7E3 9.5E3 9.3E3 9.1E3 8.6E3 8.4E3 8.1E3 8.0E3 7.8E3 7.6E3 7.6E3 7.5E3 7.3E3 7.1E3 7.0E3 6.7E3 6.7E3 6.6E3 6.5E3 6.5E3 6.2E3 6.2E3 6.0E3 5.9E3 5.9E3 5.8E3 5.7E3 5.7E3 5.7E3 5.7E3 5.4E3 5.4E3 5.4E3 5.4E3 5.4E3
1.67 1.67 1.66 1.65 1.65 1.64 1.65 1.63 1.63 1.62 1.61 1.61 1.61 1.61 1.6 1.6 1.6 1.59 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58
6.0E2 5.4E2 4.9E2 4.2E2 3.8E2 3.4E2 3.0E2 2.8E2 2.4E2 2.2E2 1.9E2 1.8E2 1.6E2 1.4E2 1.3E2 1.1E2 1.0E2 1.0E2 9.3E3 8.8E3 8.1E3 8.0E3 7.6E3 7.4E3 7.0E3 6.9E3 6.7E3 6.5E3 6.3E3 6.2E3 6.1E3 6.0E3 5.8E3 5.7E3 5.6E3 5.5E3 5.4E3 5.3E3 5.3E3 5.2E3 5.0E3 4.9E3 4.8E3 4.8E3 4.7E3 4.6E3 4.6E3 4.5E3 4.4E3 4.4E3 4.3E3 4.2E3 4.2E3 4.0E3 4.0E3 4.0E3 4.0E3 4.0E3 4.0E3 4.0E3 4.0E3 4.0E3 3.9E3 3.9E3
1.67 1.66 1.66 1.66 1.65 1.64 1.64 1.63 1.62 1.62 1.62 1.62 1.61 1.61 1.61 1.60 1.60 1.60 1.59 1.59 1.59 1.59 1.59 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.58 1.59 1.59 1.59 1.57 1.57 1.57 1.57 1.57 1.57 1.57 1.57 1.57 1.57
3.6E2 3.3E2 3.0E2 2.7E2 2.4E2 2.1E2 1.9E2 1.7E2 1.6E2 1.5E2 1.4E2 1.2E2 1.1E2 1.0E2 9.3E3 8.6E3 7.8E3 7.2E3 6.6E3 6.2E3 6.0E3 5.7E3 5.4E3 5.2E3 5.1E3 5.0E3 4.9E3 4.7E3 4.6E3 4.5E3 4.4E3 4.3E3 4.2E3 4.1E3 4.0E3 3.9E3 3.9E3 3.8E3 3.7E3 3.6E3 3.6E3 3.5E3 3.5E3 3.5E3 3.4E3 3.3E3 3.2E3 3.1E3 3.1E3 3.0E3 3.0E3 3.0E3 2.9E3 2.9E3 2.8E3 2.8E3 2.8E3 2.8E3 2.8E3 2.8E3 2.8E3 2.8E3 2.8E3 2.7E3
0.08 0.06 0.04 0.02 0.00 300
400
500
600
700
800
900
1000
1100
wavelength (nm) Fig. 9. Tholins films k values compared with Titan’s aerosols deduced from Voyager and Cassini/VIMS observations.
et al., 1992) and imaginary indices obtained for TF2, TF5, TF10 are relatively close to each other. Moreover, we remind that, in continuous working conditions, the CH4 amount into the plasma discharge is half the injected one (Sciamma-O’Brien et al., 2010). The methane consumption is certainly different from a plasma setup to another. The initial methane amount should therefore not be considered as an absolute indicator of the gas mixture in which aerosols are produced. Rannou et al. (2010) values are deduced from DISR data (Tomasko et al., 2008), in low altitudes (<140 m), using a model of scattering by fractal aggregates. k values so obtained are particularly high. These high k values obtained from Titan’s observations are quite in disagreement with aliphatic rich aerosols, whereas a large aliphatic content was suggested by the IR spectra from CIRS and VIMS measurements (Gautier et al., 2012). This apparent disagreement could possibly be explained by a difference in the chemical composition of Titan aerosols in low and high altitudes.
6. Conclusion All our measurements have been done with the same physical experimental conditions (pressure and temperature) while Imanaka et al. (2004) changed the pressure for a given methane percentage. So in this work we characterize the influence of the methane concentration on optical indices of tholins mimicking Titan’s aerosols and produced by a RF plasma discharge. An important decrease of the imaginary optical index, when the CH4 concentration increases, has been demonstrated. Mid-Infrared spectra of thin tholins films produced in different CH4 concentrations allows to connect this k variation to nitrogen-rich polymers preferably produced in lower methane percentages. In Titan’s atmosphere, the CH4 concentration varies with altitude and was varied during the evolution of this atmosphere. In this context, the study presented in this paper is very important to model albedo and then photochemical variations both in space and in time. Nevertheless, we must keep in mind that the initial amount of CH4 injected in the reactor is not the concentration in the continuous working conditions when tholins are produced. In fact, Sciamma et al. (2010) have shown that the amount of CH4 is approximately half of the injected one. Comparison of our results to data deduced from Titan’s observations raises a possible inconsistency between two observations: the mid-IR VIMS and CIRS signatures are in agreement with an aerosol dominated by aliphatic carbon content, whereas the DISR-derived important visible absorption seems to be incompatible with such an important aliphatic content.
370 380 390 400 410 420 430 440 450 460 470 480 490 500 510 520 530 540 550 560 570 580 590 600 610 620 630 640 650 660 670 680 690 700 710 720 730 740 750 760 770 780 790 800 810 820 830 840 850 860 870 880 890 900 910 920 930 940 950 960 970 980 990 1000
2%
5%
10%
Acknowledgments This work was financially supported by CNRS (PNP, ANR09-JCJC-0038 contract). We thank Dr. J.M. Coanga for his cooperation in ellipsometric measurements.
A. Mahjoub et al. / Icarus 221 (2012) 670–677
We thank Dr. P. Pernot for tholins photos and Dr. Dumas for given us the opportunity to use SMIS facility to record Mid-Infrared spectra. The authors are grateful to DGCIS of French Government, Region Ile de France and Conseil Général des Yvelines, CG78 for funding of M-2000V spectroscopic ellipsometer from J.A. Woollam Co. through FUI projects MENKAR and AUDACE from Competitive Cluster MOVEO. Appendix A See Table A1. References Alcouffe, G. et al., 2010. Capacitively Coupled Plasma used to simulate Titan’s atmospheric chemistry. Plasma Sources Sci. Technol. 19, 015008. Atreya, S.K. et al., 2006. Titan’s methane cycle. Planet. Space Sci. 54, 1177–1187. Azzam, R.M.A., Bashara, N.M., 1989. Ellipsometry and Polarized Light. NorthHolland. Berndt, J., Kovacevic, E., Stefanovic, I., Boufendi, L., 2009. J. Appl. Phys. 106, 063309. Carrasco, N. et al., 2009. Chemical characterization of Titan’s Tholins: Solubility, morphology and molecular structure revisited. J. Phys. Chem. A 113, 11195– 11203. Dahoo, P.R. et al., 2004. Evidence by spectroscopic ellipsometry of optical property change in pulsed laser deposited NiO films when heated in air at Neel temperature. Appl. Phys. A: Mater. Sci. Process. 79, 1439–1443. Derenne, S., Coelho, C., Anquetil, C., Szopa, C., Rahman, A.S., McMillan, P.F., Cora, F., Pickard, C.J., Quirico, E., Bonhomme, C., 2012. New insights into the structure and chemistry of Titan tholins via 13C and 15N solid state nuclear magnetic resonance spectroscopy. Icarus 221, 844–853. Dumas, P., Polack, B., Largarde, B., Chubar, O., Giogetta, J.L., Lefrançois, S., 2006. Synchrotron infrared microscopy at the French synchrotron facility SOLEIL. Infrared Phys. Technol. 49, 152–160. Franta, D. et al., 2005. Optical properties of NiO thin films prepared by pulsed laser deposition technique. Appl. Surf. Sci. 244, 420–426. Fujiwara, H., 2007. Spectroscopic Ellipsometry: Principles and Applications. Marusen, Tokyo. Gautier, T., Carrasco, N., Mahjoub, A., Vinatier, S., Giuliani, A., Szopa, C., Anderson, C.M., Correia, J.J., Dumas, P., Cernogora, G., 2012. Mid- and far-infrared absorption spectroscopy of Titan’s aerosols analogues. Icarus 221, 320–327. Hadamcik, E., Renard, J.-B., Alcouffe, G., Cernogora, G., Levasseur-Regourd, A.C., Szopa, C., 2009. Laboratory light scattering measurements with Titan’s aerosols analogues produced by a dusty plasma. Planet. Space Sci. 57, 1631–1641. Hasenkopf, C.A., Beaver, M.R., Trainer, M.G., Dewitt, H.L., Freedman, M.A., Toon, O.B., McKay, C.P., Tolbert, M.A., 2010. Optical properties of Titan and early Earth haze laboratory analogs in the mid-visible. Icarus 207, 903–913. Hébrard, E., Dobrijevic, M., Bélnilin, Y., Raulin, F., 2007. Photochemical kinetic uncertainties in modeling Titan’s atmosphere: First consequences. Planet. Space Sci. 55, 1470–1479. Herzinger, C.M., Johs, B., McGahan, W.A., Woollam, J.A., Paulson, W., 1998. Ellipsometric determination of optical constants for silicon and thermally grown silicon dioxide via a multi-sample, multi-wavelength, multi-angle investigation. J. Appl. Phys. 83, 3323. Horvath, G. et al., 2009. Corona discharge experiments in admixtures of N2 and CH4: A laboratory simulation of Titan’s atmosphere. Plasma Sources Sci. Technol. 18, 034016.
677
Horvath, G., Mason, N.J., Polachova, L., Zahoran, M., Moravsky, L., Matejcik, S., 2010. Packed bed DBD discharge experiments in admixtures of N2 and CH4. Plasma Chem. Plasma Process. 30, 565–577. Imanaka, H., Khare, B.N., Elsila, J.E., Bakes, E.L.O., McKay, C.P., Cruikshank, D.P., Sugita, S., Matsui, T., Zare, R.N., 2004. Laboratory experiments of Titan tholin formed in cold plasma at various pressures: Implications for nitrogencontaining polycyclic aromatic compounds in Titan haze. Icarus 168, 344–366. Jellison, G.E., Modine Jr., F.A., 1996. Parameterization of the optical functions of amorphous materials in the interband region. Appl. Phys. Lett. 69, 371–373. Khare, B.N., Sagan, C., Arakawa, E.T., Suits, F., Callcott, T.A., Williams, M.W., 1984. Optical constants of organic tholins produced in a simulated titanian atmosphere: From X-ray to microwave frequencies. Icarus 60, 127–137. Kramers, H.A., 1927. La diffusion de la lumiere par les atomes, AttiCong. Intern. Fisica, (Transactions of Volta centenary congress). Como 2, 545–557. Kronig, R.de L., 1926. On the theory of the dispersion of X-rays. J. Opt. Soc. Am. 12, 547–557. Loutete-Dangui, E.D., Varret, F., Codjovi, E., Dahoo, P.R., Boukheddaden, K., 2007. Thermal spin transition in Fe[NH3]Br2 investigated by spectroscopic ellipsometry. Phys. Rev. B 75, 184425-1–184425-7. McKay, C.P., Pollack, J.B., Courtin, R., 1989. The thermal structure of Titan’s atmosphere. Icarus 80, 23–53. Pernot, P., Carrasco, N., Thissen, R., Schmitz-Afonso, I., 2010. Tholinomics-chemical analysis of nitrogen-rich polymers. Anal. Chem. 82, 1371–1380. Quirico, E., Montagnac, G., Lees, V., McMillan, P.F., Szopa, C., Cernogora, G., Rouzaud, J.-N., Simon, P., Bernard, J.-M., Coll, P., 2008. New experimental constraints on the composition and structure of tholins. Icarus 198, 218–231. Rages, K., Pollack, J.B., 1980. Titan aerosols: Optical proprieties and vertical distribution. Icarus 41, 119–130. Ramirez, S.I., Coll, P., da Silva, A., Navarro-González, R., Lafait, J., Raulin, F., 2002. Complex refractive index of Titan’s aerosol analogues in the 200–900 nm domain. Icarus 156, 515–529. Rannou, P., Cours, T., Le Mouélic, S., Rodriguez, S., Sotin, C., Drossart, P., Brown, R., 2010. Titan haze distribution and optical properties retrieved from recent observations. Icarus 208, 850–867. Sagan, C., Thompson, W.R., Khare, B.N., 1992. Titan: A laboratory for prebiological organic chemistry. Acc. Chem. Res. 25, 286–292. Samuelson, R.E., 1983. Radiative equilibrium model of Titan’s. Icarus 53, 364–387. Sciamma-O’Brien, E., Carrasco, N., Szopa, C., Buch, A., Cernogora, G., 2010. Titan’s atmosphere: An optimal gas mixture for aerosol formation? Icarus 209, 704– 714. Sciamma-O’Brien, E., Dahoo, P.R., Hadamcik, E., Carrasco, N., Quirico, E., Szopa, C., Cernogora, G., 2012. Optical constants from 370 nm to 900 nm of Titan tholins produced in a low pressure RF plasma discharge. Icarus 218, 356–363. Szopa, C., Cernogora, G., Boufendi, L., Correia, J.-J., Coll, P., 2006. PAMPRE: A dusty plasma experiment for Titan’s tholins production and study. Planet. Space Sci. 54, 394–404. Tobie, G., Lunine, J., Sotin, C., 2006. Episodic outgassing as the origin of atmospheric methane on Titan. Nature 440 (7080), 61–64. Tomasko, M. et al., 2008. A model of Titan’s aerosols based on measurements made inside the atmosphere. Planet. Space Sci. 56, 669–707. Tran, B.N., Joseph, J.C., Ferris, J.P., Persan, P.D., Chera, J.J., 2003. Simulation of haze formation using a photochemical flow reactor. The optical constants of the polymers. Icarus 165, 379–390. Vuitton, V., Tran, B.N., Persans, P.D., Ferris, J.P., 2009. Determination of the complex refractive indices of Titan haze analogs using photothermal deflection spectroscopy. Icarus 203, 663–671. Waite, J.H. et al., 2007. The process of tholins formation in Titan’s upper atmosphere. Science 316, 870–875. Weber, J.W., Hansen, T.A.R., van de Sanden, M.C.M., Engeln, R., 2009. B-Spline parametrisation of the dielectric function applied to spectroscopic ellipsometry on amorphous carbon. J. Appl. Phys. 106, 3460–3478.