Cage occupancy and structural changes during hydrate formation from initial stages to resulting hydrate phase

https://doi.org/10.1016/j.saa.2013.06.065Get rights and content

Highlights

  • We synthesized mixed gas hydrates from ice and different gas mixtures.

  • The gas hydrate formation process was investigated in situ via Raman spectroscopy.

  • All systems formed pentagonal dodecahedrons occupied with methane as a first step.

  • The initial cage occupancy and composition differs from final hydrate phase.

Abstract

Hydrate formation processes and kinetics are still not sufficiently understood on a molecular level based on experimental data. In particular, the cavity formation and occupancy during the initial formation and growth processes of mixed gas hydrates are rarely investigated. In this study, we present the results of our time-depending Raman spectroscopic measurements during the formation of hydrates from ice and gases or gas mixtures such as CH4, CH4–CO2, CH4–H2S, CH4–C3H8, CH4-iso-C4H10, and CH4-neo-C5H12 at constant pressure and temperature conditions and constant composition of the feed gas phase. All investigated systems in this study show the incorporation of CH4 into the 512 cavities as first step in the initial stages of hydrate formation. Furthermore, the results imply that the initial hydrate phases differ from the resulting hydrate phase having reached a steady state regarding the occupancy and ratio of the small and large cavities of the hydrate.

Introduction

Clathrate hydrates are formed from natural gases in nature as well as in industrial processes. In particular, for the latter the hydrate formation is not always desired. In any case the understanding of the formation process and its kinetics might be helpful for predictions regarding the expected amount of hydrates formed under defined conditions. Several efforts in the past give some insights regarding the formation kinetics. An easy but efficient method is the determination of pressure changes related to hydrate formation over time. These kinds of experiments have been performed with pure gases and gas mixtures and provide information about the transformation of ice/water and gas into gas hydrates over time and about the induction time for the studied system under defined conditions [1]. During the last years gas hydrate research has increasingly focused on a molecular level including in situ techniques such as Raman spectroscopy, X-ray diffraction and NMR spectroscopy. Recently, Luzi et al. presented data from kinetic studies on different mixed hydrates using X-ray diffraction, indicating that the reaction mechanism is different for each hydrate system and depends on the guest molecule [2]. Unfortunately, such experiments mentioned above do not provide data regarding the composition of the hydrate phase or the cage occupancies. Subramanian and Sloan as well as Uchida et al. received more details when they performed in situ observations of CH4 hydrate formation with Raman spectroscopy [3], [4]. These experiments lead to the conclusion that the 512 cavities of structure I are formed preferentially at the initial stages of hydrate formation, indicating that the formation of the 51262 cavities may be rate-limiting in structure I CH4 hydrate formation. Another tool to study CH4 hydrate formation kinetics is NMR spectroscopy: Susilo et al. studied the conversion from fresh ice and “memory” ice to CH4 hydrate in presence of a non-aqueous liquid with NMR measurements [5]. Detailed information about the local water structure before and during CH4 hydrate formation could be provided with neutron diffraction experiments by Thompson et al. [6]. It could be shown that the presence of hydrate crystallites affects the local water structure.

Natural gas usually contains predominantly CH4 but it may also contain small amounts of other gases such as light hydrocarbons, H2S or CO2. Therefore, the formation kinetics of mixed hydrates is of special interest. Kini et al. performed 13C NMR studies during structure II hydrate formation from a gas mixture containing CH4 and C3H8 and ice [7]. They observed that the large cages (51264) filled with C3H8 form twice as fast as small cages (512) occupied with CH4. Unfortunately this result differs from that of Fleyfel et al. who concluded that the small cavities are filled more rapidly [8]. This conclusion is based on their 13C NMR studies of the formation of structure II hydrates from a gas mixture containing CH4 and C3H8 and liquid water. Due to the fact that the literature data are not consistent and limited regarding the studied mixtures containing CH4 and an additional gas component, we performed time-resolved Raman spectroscopic measurements in situ on hydrate formation from ice and pure CH4 as well as different gas mixtures such as CH4–CO2, CH4–H2S, CH4–C3H8, CH4-iso-C4H10 and CH4-neo-C5H12 at constant pressure and temperature conditions and constant composition of the feed gas phase.

Section snippets

Materials and methods

The hydrate formation process was studied starting from freshly prepared ice. The ice was generated from deionized water that was frozen in a liquid nitrogen bath. The ice was powdered in a 6750 freezer mill (Spex CertiPrep) that was also cooled with liquid nitrogen. The diameter of these ice particles was determined with scanning electron microscopy (SEM ULTRA plus) and lies in between 10 and 20 μm. Therefore, the ice appears rather foamy and may contain occluded gas. Fig. 1 shows SEM images of

Results and discussion

The first experiments performed to investigate hydrate formation kinetics and growth were time-resolved Raman spectroscopic measurements on pure CH4 hydrates. Similar experiments have been performed before by Subramanian and Sloan for CH4 and liquid water [3]. This literature data in addition to our experimental results from investigations on the pure CH4 hydrate can be used as basis for subsequent interpretation and discussion regarding the formation of hydrates synthesized from gas mixtures.

Conclusion

Time-resolved investigations on hydrate formation with Raman spectroscopy were performed on hydrates formed from ice and CH4, CH4–H2S, CH4–CO2, CH4–C3H8, CH4-iso-C4H10 and CH4-neo-C5H12. All investigated systems showed at the initial stage of the formation process the incorporation of CH4 into the solid phase in a manner that reminds on the formation of small 512 cavities which could be proofed with a band in the Raman spectra at 2915 cm−1. The preferred formation of small 512 cavities during

References (33)

  • P. Skovborg et al.

    Chem. Eng. Sci.

    (1994)
  • M. Luzi et al.

    J. Chem. Thermodyn.

    (2012)
  • S. Subramanian et al.

    Fluid Phase Equilib.

    (1999)
  • E.A.J. Burke

    Lithos

    (2001)
  • J.H. Schachtschneider et al.

    Spectrochim. Acta

    (1963)
  • N.G. Mirkin et al.

    J. Mol. Struct.

    (2000)
  • D. Kashchiev et al.

    J. Cryst. Growth

    (2002)
  • J.M. Schicks et al.

    Spectrochim. Acta A

    (2010)
  • T. Uchida et al.

    Ann. NY Acad. Sci.

    (2000)
  • R. Susilo et al.

    J. Phys. Chem. B

    (2006)
  • H. Thompson et al.

    J. Chem. Phys.

    (2006)
  • R.A. Kini et al.

    J. Phys. Chem. A

    (2004)
  • F. Fleyfel et al.

    Ann. NY Acad. Sci.

    (1994)
  • J.M. Schicks et al.

    Angew. Chem. Int. Ed.

    (2004)
  • J.M. Schicks et al.

    J. Phys. Chem. B

    (2006)
  • J.M. Schicks, H. Lu, J.A. Ripmeester, M. Ziemann, in: Proceedings of the 6th International Conference on Gas Hydrates,...
  • Cited by (45)

    • Experimental investigation of hydrate formation kinetics and microscopic properties by a synthesized ternary gas mixture with combination additives

      2022, Energy
      Citation Excerpt :

      Raman band positions of C–O stretching vibrations of CO2 hydrate was between at 800–1600 cm−1. The C–H symmetric stretching region of CH4 gas molecules in hydrate was located at 2700–3050 cm−1 [44,45]. Fig. 7 shows the Raman spectra in the ternary gas mixture hydrate with the volume ratio of CP/SDS concentration of 0.308 and 0.154, respectively.

    View all citing articles on Scopus
    View full text