
Download
XML
0.00MB

Download
PDF
0.00MB

Read Online
HTML
0.00MB
Abstract
In this (Paper I) and the companion paper (Paper II; R. May, R. Smith, and B. Kay, J. Chem. Phys.138, 104502 (Year: 2013)10.1063/1.4793312), we investigate the mechanisms for the release of trapped gases from underneath amorphous solid water (ASW) films. In prior work, we reported the episodic release of trapped gases in concert with the crystallization of ASW, a phenomenon that we termed the “molecular volcano.” The observed abrupt desorption is due to the formation of cracks that span the film to form a connected pathway for release. In this paper, we utilize the “molecular volcano” desorption peak to characterize the formation of crystallizationinduced cracks. We find that the crack length distribution is independent of the trapped gas (Ar, Kr, Xe, CH_{4}, N_{2}, O_{2}, or CO). Selective placement of the inert gas layer is used to show that cracks form near the top of the film and propagate downward into the film. Isothermal experiments reveal that, after some induction time, cracks propagate linearly in time with an Arrhenius dependent velocity corresponding to an activation energy of 54 kJ/mol. This value is consistent with the crystallization growth rates reported by others and establishes a direct connection between crystallization growth rate and the crack propagation rate. A twostep model in which nucleation and crystallization occurs in an induction zone near the top of the film followed by the propagation of a crystallization/crack front into the film is in good agreement with the temperature programmed desorption results.
This work was supported by the U.S. Department of Energy (DOE), Office of Basic Energy Sciences, Division of Chemical Sciences, Geosciences, and Biosciences. The research was performed using EMSL, a national scientific user facility sponsored by DOE's Office of Biological and Environmental Research and located at Pacific Northwest National Laboratory, which is operated by Battelle for the U.S. Department of Energy under Contract No. DEAC0576RL01830.
I. INTRODUCTION
II. EXPERIMENTAL
III. RESULTS AND DISCUSSION
A. Characterization of crystallizationinduced crack formation in ASW using the molecular volcanodesorption peak
B. “Topdown” crystallization and crack propagation in amorphous solid water
C. Topdown crystallizationinduced crack propagation kinetics
D. Coupled induction time and “topdown” crystallization/crack front propagation kinetic model
IV. CONCLUSIONS
Key Topics
 Desorption
 68.0
 Crystallization
 50.0
 Volcanoes
 49.0
 Activation energies
 17.0
 Thin films
 15.0
B01D9/00
Figures
Temperature programmed desorption (TPD) spectra of 1 ML of Ar, Kr, and Xe desorbing directly from graphene (dashed lines) and covered by 300 ML of amorphous solid water (solid lines) at a heating rate of 1 K/s. The top panel displays a representative TPD spectrum for the ASW overlayer.
Click to view
Temperature programmed desorption (TPD) spectra of 1 ML of Ar, Kr, and Xe desorbing directly from graphene (dashed lines) and covered by 300 ML of amorphous solid water (solid lines) at a heating rate of 1 K/s. The top panel displays a representative TPD spectrum for the ASW overlayer.
TPD spectra 1 ML of Ar covered by ASW overlayer thicknesses from 25 to 800 ML. The heating rate was 1 K/s. The Ar desorption shifts from the “volcano” peak to the “trapped” peak with increasing overlayer coverage.
Click to view
TPD spectra 1 ML of Ar covered by ASW overlayer thicknesses from 25 to 800 ML. The heating rate was 1 K/s. The Ar desorption shifts from the “volcano” peak to the “trapped” peak with increasing overlayer coverage.
(a) The fraction of 1 ML of Ar (open circles) that remains “trapped” as a function of overlayer thickness at a heating rate of 1 K/s. The line through the experimental data is a fit using Eq. (1) . The other line is the derivative, dF/dL, of the fit line. The derivative curve yields the distribution of vertical crack lengths. (b) The vertical crack length distributions for Ar, Kr, Xe, CH_{4}, N_{2}, O_{2}, and CO obtained from the derivative of the trapped fraction versus overlayer thickness for the individual species as is demonstrated in (a). The vertical dashed line is the mean crack length for all of the gases.
Click to view
(a) The fraction of 1 ML of Ar (open circles) that remains “trapped” as a function of overlayer thickness at a heating rate of 1 K/s. The line through the experimental data is a fit using Eq. (1) . The other line is the derivative, dF/dL, of the fit line. The derivative curve yields the distribution of vertical crack lengths. (b) The vertical crack length distributions for Ar, Kr, Xe, CH_{4}, N_{2}, O_{2}, and CO obtained from the derivative of the trapped fraction versus overlayer thickness for the individual species as is demonstrated in (a). The vertical dashed line is the mean crack length for all of the gases.
(a) The fraction of 1 ML of Ar that remains “trapped” as a function of overlayer thickness for a series of heating rates. The lines are fits to the trapped fraction versus overlayer thickness similar to that in Fig. 3(a) . (b) Derivative, dF/dL, of the lines in (a) indicating the vertical crack length distribution.
Click to view
(a) The fraction of 1 ML of Ar that remains “trapped” as a function of overlayer thickness for a series of heating rates. The lines are fits to the trapped fraction versus overlayer thickness similar to that in Fig. 3(a) . (b) Derivative, dF/dL, of the lines in (a) indicating the vertical crack length distribution.
Comparison of two methods for varying the amount of ASW on top of 1 ML Ar. In the first method (squares), 1 ML Ar is fixed beneath increasing amounts of ASW while in the second method (circles) 1 ML Ar is moved through a fixed 600 ML ASW film. (a) Comparison of the volcano peak temperature, T _{ V }, between the two methods as a function of the thickness of ASW on top. For both methods, T _{ V } is nearly identical for a given overlayer thickness. (b) Comparison of the trapped fraction between the two methods as a function of the thickness of ASW on top. At a given point the trapped fraction is larger when Ar is sandwiched between layers of ASW.
Click to view
Comparison of two methods for varying the amount of ASW on top of 1 ML Ar. In the first method (squares), 1 ML Ar is fixed beneath increasing amounts of ASW while in the second method (circles) 1 ML Ar is moved through a fixed 600 ML ASW film. (a) Comparison of the volcano peak temperature, T _{ V }, between the two methods as a function of the thickness of ASW on top. For both methods, T _{ V } is nearly identical for a given overlayer thickness. (b) Comparison of the trapped fraction between the two methods as a function of the thickness of ASW on top. At a given point the trapped fraction is larger when Ar is sandwiched between layers of ASW.
(a) Volcano peak temperatures, T _{ v }, for Ar (squares) placed at the bottom of and for O_{2} (circles) placed at various levels in a 300 ML ASW. The Ar volcano peak temperature remains constant while the O_{2} peak moves to lower temperature as it is placed higher in the ASW. (b) The corresponding trapped fraction for Ar (squares) placed at the bottom of and for O_{2} (circles) placed at various levels in a 300 ML ASW.
Click to view
(a) Volcano peak temperatures, T _{ v }, for Ar (squares) placed at the bottom of and for O_{2} (circles) placed at various levels in a 300 ML ASW. The Ar volcano peak temperature remains constant while the O_{2} peak moves to lower temperature as it is placed higher in the ASW. (b) The corresponding trapped fraction for Ar (squares) placed at the bottom of and for O_{2} (circles) placed at various levels in a 300 ML ASW.
(a) Desorption spectra (solid lines) of 1 ML of Ar at an isothermal temperature of 154 K with varying ASW overlayer thickness. The dashed line is a representative ASW spectrum that was nearly the same in all of the experiments independent of the overlayer thickness. (b) A plot of the volcano peak time, t _{ V }, (symbols) versus ASW overlayer thickness at a variety of isothermal temperatures. The lines are linear fits to the data above 100 ML.
Click to view
(a) Desorption spectra (solid lines) of 1 ML of Ar at an isothermal temperature of 154 K with varying ASW overlayer thickness. The dashed line is a representative ASW spectrum that was nearly the same in all of the experiments independent of the overlayer thickness. (b) A plot of the volcano peak time, t _{ V }, (symbols) versus ASW overlayer thickness at a variety of isothermal temperatures. The lines are linear fits to the data above 100 ML.
(a) Arrhenius plot of the apparent crack velocity (□) obtained from the slopes of linear fits of the volcano time versus overlayer thickness for the isothermal measurements in Fig. 7 and others. The fit parameters were v _{ o } = 4.3 × 10^{19} ML/s and E _{ a } = 54 kJ/mol. (Inset) Comparison of apparent crack velocity data with ASW crystallization growth rates from Dohnalek et al. ^{ 33 } (■), Backus et al. ^{ 32 } (★), and Safarik and Mullins ^{ 34 } (◆). (b) Arrhenius fit to induction times form the yintercept of linear fits of the volcano time versus overlayer thickness for the isothermal measurements in Fig. 7 and others (◯). The fit parameters were v _{ o } = 5.2 × 10^{−27} s and E _{ a } = 80 kJ/mol. (Inset) Comparison of induction time data with ASW crystallization induction time from Dohnalek et al. ^{ 33 } (●). Error bars represent an estimated 10 s uncertainty in the time the isothermal temperature is reached during the heating ramp.
Click to view
(a) Arrhenius plot of the apparent crack velocity (□) obtained from the slopes of linear fits of the volcano time versus overlayer thickness for the isothermal measurements in Fig. 7 and others. The fit parameters were v _{ o } = 4.3 × 10^{19} ML/s and E _{ a } = 54 kJ/mol. (Inset) Comparison of apparent crack velocity data with ASW crystallization growth rates from Dohnalek et al. ^{ 33 } (■), Backus et al. ^{ 32 } (★), and Safarik and Mullins ^{ 34 } (◆). (b) Arrhenius fit to induction times form the yintercept of linear fits of the volcano time versus overlayer thickness for the isothermal measurements in Fig. 7 and others (◯). The fit parameters were v _{ o } = 5.2 × 10^{−27} s and E _{ a } = 80 kJ/mol. (Inset) Comparison of induction time data with ASW crystallization induction time from Dohnalek et al. ^{ 33 } (●). Error bars represent an estimated 10 s uncertainty in the time the isothermal temperature is reached during the heating ramp.
(a) Volcano peak temperature, T _{ v }, as a function of ASW overlayer thickness, L, for a series of heating rates from 0.1 to 2 K/s. (b) Volcano peak temperature, T _{ v }, versus the overlayer thickness scaled by the heating rate, βL. The thickness rescaling collapses the T _{ v } data onto a single curve. (c) Arrhenius plot of the quantity, , a quantity that is shown in Eq. (5) to be proportional to the crack velocity, v(T _{ V }). The nonlinearity of the plot indicates that the activation energy is temperature and heating rate dependent in contrast to the isothermal results in Figs. 7 and 8 .
Click to view
(a) Volcano peak temperature, T _{ v }, as a function of ASW overlayer thickness, L, for a series of heating rates from 0.1 to 2 K/s. (b) Volcano peak temperature, T _{ v }, versus the overlayer thickness scaled by the heating rate, βL. The thickness rescaling collapses the T _{ v } data onto a single curve. (c) Arrhenius plot of the quantity, , a quantity that is shown in Eq. (5) to be proportional to the crack velocity, v(T _{ V }). The nonlinearity of the plot indicates that the activation energy is temperature and heating rate dependent in contrast to the isothermal results in Figs. 7 and 8 .
Coupled induction and crystallization growth front propagation model. ASW crystallization begins at or near the outer surface, then after an induction period, the crystallization front propagates downward leading to the formation of cracks. Finally, the cracks reach the bottom of the ASW releasing the covered gas.
Click to view
Coupled induction and crystallization growth front propagation model. ASW crystallization begins at or near the outer surface, then after an induction period, the crystallization front propagates downward leading to the formation of cracks. Finally, the cracks reach the bottom of the ASW releasing the covered gas.
(a) Arrhenius plot of the crystallization/crack front velocities calculated using Eq. (9) , which was derived using the coupled induction and front propagation model. The dashed line is the temperature dependent velocity obtained from isothermal experiments. (b) Arrhenius plot of experimental crystallization/crack front velocities calculated using Eq. (5) . The model calculated and experimental velocities are qualitatively similar albeit shifted by ∼2.0 K.
Click to view
(a) Arrhenius plot of the crystallization/crack front velocities calculated using Eq. (9) , which was derived using the coupled induction and front propagation model. The dashed line is the temperature dependent velocity obtained from isothermal experiments. (b) Arrhenius plot of experimental crystallization/crack front velocities calculated using Eq. (5) . The model calculated and experimental velocities are qualitatively similar albeit shifted by ∼2.0 K.
Crystallization/crack front propagation activation energy as a function of ASW overlayer thickness determined from TPD experiments (squares) and from the coupled induction and front propagation model (solid line). The horizontal dashed lines demark the activation energies for induction (80 kJ/mol) and crack front propagation (54 kJ/mol) determined from the isothermal experiments in Fig. 8 .
Click to view
Crystallization/crack front propagation activation energy as a function of ASW overlayer thickness determined from TPD experiments (squares) and from the coupled induction and front propagation model (solid line). The horizontal dashed lines demark the activation energies for induction (80 kJ/mol) and crack front propagation (54 kJ/mol) determined from the isothermal experiments in Fig. 8 .
Article metrics loading...
Abstract
In this (Paper I) and the companion paper (Paper II; R. May, R. Smith, and B. Kay, J. Chem. Phys.138, 104502 (Year: 2013)10.1063/1.4793312), we investigate the mechanisms for the release of trapped gases from underneath amorphous solid water (ASW) films. In prior work, we reported the episodic release of trapped gases in concert with the crystallization of ASW, a phenomenon that we termed the “molecular volcano.” The observed abrupt desorption is due to the formation of cracks that span the film to form a connected pathway for release. In this paper, we utilize the “molecular volcano” desorption peak to characterize the formation of crystallizationinduced cracks. We find that the crack length distribution is independent of the trapped gas (Ar, Kr, Xe, CH_{4}, N_{2}, O_{2}, or CO). Selective placement of the inert gas layer is used to show that cracks form near the top of the film and propagate downward into the film. Isothermal experiments reveal that, after some induction time, cracks propagate linearly in time with an Arrhenius dependent velocity corresponding to an activation energy of 54 kJ/mol. This value is consistent with the crystallization growth rates reported by others and establishes a direct connection between crystallization growth rate and the crack propagation rate. A twostep model in which nucleation and crystallization occurs in an induction zone near the top of the film followed by the propagation of a crystallization/crack front into the film is in good agreement with the temperature programmed desorption results.
Full text loading...
Commenting has been disabled for this content