10:45 AM - 12:15 PM
[MIS14-P03] Development of a graphene liquid cell for direct TEM observation of ice nucleation
Keywords:Ice, Nucleation, TEM, Graphene
To fully understand the nature of ice crystallization, it is important to focus on the early stage of the process, starting from the formation of ice nanoparticles. According to classical nucleation theory, the interfacial energy, which is a characteristic parameter of ice particles in water, can be derived from the nucleation energy barrier with a certain degree of supercooling ΔT. This barrier is mainly influenced by the Gibbs free energy of the evolving solid nuclei, which itself could also be in metastable phases according to Ostwald's step rule. In addition, a two-step growth mechanism starting from amorphous state has been discovered in several materials [1]. Therefore, it is expected to visualize the dynamics of formation and growth of ice nanoparticles at solid-liquid interface with controlled ΔT. It is still not revealed experimentally due to technical constraints, remaining in a computer simulation [2].
For direct observation of the nucleation processes, transmission electron microscopy (TEM) has some advantages due to its sufficiently high spatial and temporal resolution. Since TEM uses electron as a probe, it is necessary to keep the specimen chamber in high vacuum condition. Therefore, we need to encapsulate a solution to prevent evaporation. As a conventional method, thin SiNx films have been used for sealing. This technique is effective in capturing crystal growth of solutes that have relatively high atomic number in aqueous solution [3]. In the case of ice crystal growth from pure water, Tai et al. succeeded in observing the Ic phase added to the normal Ih phase under the ice growth [4]. It becomes difficult to obtain clear contrast images due to the thick SiNx films and lower atomic number of oxygen in water than Si in the films. Therefore, it has been limited to observe the ice growth process after the nucleation process.
Graphene, a periodic hexagonal network of C atoms in two dimensional, has been noticed as an alternative material. Yak et al. recently developed a graphene liquid cell (GLC) by facing two graphene sheets with each other [5]. Graphene itself have strong hydrophobicity, but remaining liquid pockets is formed induced by strong van der Waals interaction between the sheets under water expelling process. The high tolerance and transparency of graphene to electrons allow us to observe any liquid pockets in nm-order spatial resolution. However, GLCs are often contaminated during the fabrication process. Here, we investigated several methods to fabricate GLCs and minimize contamination. We also attempted cryo-TEM observation using the fabricated GLCs.
By facing the free-standing graphene with that supported by a holey carbon TEM grid, we finally succeeded in trapping water in GLC with minimal contamination. Energy dispersive X-ray spectroscopy confirmed the absence of contamination from etchant of graphene, which was a major obstacle to observing ice nucleation. Suitable conditions such as the smoothness of the graphene and the graphene transfer methods would be discussed.
A cooling experiment was performed on the GLC using a cryo-TEM holder that allows the sample to be cooled to cryogenic temperatures. At 80 K, several contrasts were identified in the TEM image. The presence of oxygen at the spots was confirmed by energy-filtered TEM, indicating that the water was successfully encapsulated between two sheets of graphene. In addition, the electron diffraction spots of these were not appeared. These results suggest that the water was frozen as amorphous ice. The day will include a detailed discussion of the ice conditions in the GLC during the cooling experiment and a roadmap for observing ice nucleation.
[1] M. H. Nielsen et al., Science. 345, 1158–1162 (2014).
[2] K. K. Tanaka et al., Phys Rev E. 96, 022804 (2017).
[3] T. Yamazaki et al., Proc Natl Acad Sci USA. 114, 2154–2159 (2017).
[4] K. Tai et al., Microsc Microanal. 20, 330–337 (2014).
[5] J. M. Yuk et al., Science. 336, 61–64 (2012).
For direct observation of the nucleation processes, transmission electron microscopy (TEM) has some advantages due to its sufficiently high spatial and temporal resolution. Since TEM uses electron as a probe, it is necessary to keep the specimen chamber in high vacuum condition. Therefore, we need to encapsulate a solution to prevent evaporation. As a conventional method, thin SiNx films have been used for sealing. This technique is effective in capturing crystal growth of solutes that have relatively high atomic number in aqueous solution [3]. In the case of ice crystal growth from pure water, Tai et al. succeeded in observing the Ic phase added to the normal Ih phase under the ice growth [4]. It becomes difficult to obtain clear contrast images due to the thick SiNx films and lower atomic number of oxygen in water than Si in the films. Therefore, it has been limited to observe the ice growth process after the nucleation process.
Graphene, a periodic hexagonal network of C atoms in two dimensional, has been noticed as an alternative material. Yak et al. recently developed a graphene liquid cell (GLC) by facing two graphene sheets with each other [5]. Graphene itself have strong hydrophobicity, but remaining liquid pockets is formed induced by strong van der Waals interaction between the sheets under water expelling process. The high tolerance and transparency of graphene to electrons allow us to observe any liquid pockets in nm-order spatial resolution. However, GLCs are often contaminated during the fabrication process. Here, we investigated several methods to fabricate GLCs and minimize contamination. We also attempted cryo-TEM observation using the fabricated GLCs.
By facing the free-standing graphene with that supported by a holey carbon TEM grid, we finally succeeded in trapping water in GLC with minimal contamination. Energy dispersive X-ray spectroscopy confirmed the absence of contamination from etchant of graphene, which was a major obstacle to observing ice nucleation. Suitable conditions such as the smoothness of the graphene and the graphene transfer methods would be discussed.
A cooling experiment was performed on the GLC using a cryo-TEM holder that allows the sample to be cooled to cryogenic temperatures. At 80 K, several contrasts were identified in the TEM image. The presence of oxygen at the spots was confirmed by energy-filtered TEM, indicating that the water was successfully encapsulated between two sheets of graphene. In addition, the electron diffraction spots of these were not appeared. These results suggest that the water was frozen as amorphous ice. The day will include a detailed discussion of the ice conditions in the GLC during the cooling experiment and a roadmap for observing ice nucleation.
[1] M. H. Nielsen et al., Science. 345, 1158–1162 (2014).
[2] K. K. Tanaka et al., Phys Rev E. 96, 022804 (2017).
[3] T. Yamazaki et al., Proc Natl Acad Sci USA. 114, 2154–2159 (2017).
[4] K. Tai et al., Microsc Microanal. 20, 330–337 (2014).
[5] J. M. Yuk et al., Science. 336, 61–64 (2012).