 HYDRATION AND DEHYDRATION OF MARS-RELEVANT CHLORIDE AND PERCHLORATE SALTS AT GALE CRATER.  K. M. Primm1, R. V. Gough1, E. G. Rivera-Valentín2,3, G. M. Martínez4, and M. A. Tolbert1. 1Cooperative Institute for Research in Environmental Sciences and Department of Chemistry and Biochemistry, University of Colorado, Boulder, CO, 80309, USA (Katherine.Primm@colorado.edu) 2Lunar and Planetary Institute, Houston, TX; 3Arecibo Observatory, Universities Space Research Association, Arecibo, PR; 4Department of Climate and Space Sciences and Engineering, University of Michigan, Ann Arbor, MI, USA.  Introduction:  Perchlorate and chloride salts have been found in several locations on Mars [1-4]. These salts are of interest because their hygroscopic nature and low eutectic temperatures allow for the possibility of liquid water on the surface of Mars today [5-6]. Many perchlorate and chloride salts can potentially absorb water vapor and turn into liquid solutions (i.e., deliquesce) under some Mars-relevant conditions, such as those at the Phoenix landing site; however, drier locations on Mars, such as Gale Crater, are less likely to permit the formation of liquid brines. At these drier locations, perchlorate could still be interacting with water vapor via hydration-dehydration cycles. This hydration and dehydration of crystalline salts could be at least partially responsible for the diurnal variations in near-surface atmospheric water vapor observed at both the Phoenix [7] and Curiosity [8,9] landing sites.  This nighttime decrease in water vapor could be due to frost formation at the surface [10]. Other potential processes include diffusion and adsorption of water in subsurface soil [8], soluble salt deliquescence, or salt hydration changes. Many salts found at these landing sites and elsewhere on Mars (sulfates, perchlorates and chlorides) have multiple potential hydration states, suggesting that phase transtions may be possible. It is important to understand if transitions between two or more hydration states of a salt can occur under Marsrelevant conditions and time scales. If so, such salts can provide a diurnal sink for atmospheric water vapor.  The possibility of humidity-induced salt hydration is also relevant to recurring slope lineae (RSL), the dark, recurring streaks that are proposed by some to be formed by liquid  [11]. Hydrated perchlorate and chloride salts were spectrally detected at several RSL locations, prompting the suggestion that the hydrated salts were likely precipitated from a liquid brine [4]. Here we question if the salts could have hydrated via a solid-solid phase transition involving the incorporation of atmopsheric H2O vapor into its crystal structure. If hydration can occur due to changes in relative humidity (RH) or temperature, then liquid water is not required to produce hydrated salts in RSL.   We perform experiments to understand the conditions needed for hydration and dehydration of five different Mars-relevant salts: NaCl, NaClO4, CaCl2, Ca(ClO4)2, and MgCl2. In order to understand the role of solid-solid salt hydration on Mars, we compare the results to the environmental conditions observed by the Curiosity rover in Gale Crater.  Experimental Methods: Aqueous solutions of NaCl, NaClO4, CaCl2, Ca(ClO4)2, and MgCl2 in HPLC grade water were nebulized using N2 gas onto a hydrophobic quartz disc. The resulting particle diameters were between 5 and 50 µm. The sample was placed into an environmental chamber within a Raman microscope [12-14]. The chamber allows for RH and temperature control (0.2% and 0.1K precision, respectively) while pure CO2 flows through the system. The sample was allowed to dry in a ~0% RH environment prior to beginning an experiment to ensure it was dry (not liquid).  The particles were analyzed by Raman spectroscopy to determine the phase of H2O. Figure 1 shows MgCl2 hydrating from 4H2O to 6H2O. Because chloride does not have a Raman signature, only the O-H stretching region of the Raman spectrum (3000-3800 cm-1) is shown in Fig. 1. The stable hydrate for MgCl2 at room temperature and ~0% RH is MgCl24H2O (red trace). As the humidity increases to 12.8% RH, a solidsolid hydration phase transition occurs from MgCl24H2O (3438 cm-1) to MgCl26H2O (3513, 3390, and 3349 cm-1) [15]. The images below the spectral series in Fig. 1 show the particle brightened at 12.8% RH as the 6H2O forms. This hydration procedure was attempted for all salts at a variety of temperatures. If hydration ocFigure 1. Hydration of a MgCl24H2O particle to MgCl26H2O is shown spectrally (upper panel) and optically (lower images). In all cases, the spectra color and the image frame color correspond to the same RH and T conditions. curred, a dehydration experiment was then performed. The RH was decreased while the temperature was increased in order to see if any spectral changes indicative of dehydration occurred.  Results: Although NaCl and NaClO4 do have stable hydrates [5,12], no hydration was observed after exposing the anhydrous salts to 65% RH for 8 hours [12,16]. In constrast, hydration was observed for CaCl2, Ca(ClO4)2, and MgCl2 salts [12,17,18].     Figure 2 shows the observed RH and temperature of hydration and dehydration of these three salts: (a) CaCl2, (b) Ca(ClO4)2, and (c) MgCl2. In each plot in Fig. 2, MSL Rover Environmental Monitoring Station (REMS) ground relative humidity (RHg) and temperature (Tg) up to sol 1527 (grey circles) and modeled subsurface conditions at four Ls values (62-cyan, 151orange, 241-gold, 330-pink) are plotted. The hydration and dehydration of MgCl2 was performed at both Earth pressure (630 Torr) and Martian pressure (5 Torr) and no discernable difference was detected. Each symbol represents a single hydration or dehydration experiment. Figure 2 shows that all three salts can hydrate by exposure to water vapor. Often only very low humidity (RH < 20%) is required. The only salt studied here that was observed to hydrate at the surface at Gale Crater is Ca(ClO4)2, which can hydrate from the anhydrous phase to Ca(ClO4)24H2O (red symbols). However, Gale Crater does not appear to be warm enough to allow for dehydration. In the case of Ca(ClO4)24H2O as well as MgCl26H2O, dehydration only occurred at 298 K despite waiting for 5-8 hours at 280 K at low RH (<0.1%). These salts will likely remain in the hydrated state throughout the year. The only salt observed to dehydrate under conditions found at Gale Crater is CaCl2; however, hydration is not predicted at the surface. The subsurface conditions at Gale Crater are quite different (colored lines in Fig. 2) and future studies will simulate these conditions to see if any candidate salts can both hydrate and dehydrate under Martian conditions. Conclusion: Even though multiple hydration states of Martian salts are possible and even predicted theoretically [17], the phase transitions are not always possible in practice and experiments must be performed to determine the feasibility of phase transitions of salts at the surface or in the shallow subsurface. We were unable to hydrate either NaCl or NaClO4 at all. Under Gale Crater conditions, hydration is possible for Ca(ClO4)2 and MgCl24H2O but unlikely for CaCl2; however, dehydration is observed only for CaCl2 under Gale Crater conditions and is unlikely to occur for Ca(ClO4)2 and MgCl24H2O unless the conditions reach 298 K. References: [1] Osterloo et al. (2008) Science 319. [2] Hecht et al. (2009) Science 325. [3] Glavin et al. (2013) JGR 118. [4] Ojha et al. (2015) Nature Geosci. 8. [5] Chevrier et al. (2009) GRL 36. [6] Marion et al. (2010) Icarus 207. [7] Zent et al. (2010) JGRE 115. [8] Savijarvi et al. (2015) JGRE 120. [9] Savijarvi et al. (2016) Icarus 265. [10] Chevrier and Rivera-Valentin (2012) GRL 39. [11] Martinez et al. (2016) Icarus 280. [12] Gough et al. (2011) EPSL 312. [13] Baustian et al. (2010) ACP 10. [14] Schill and Tolbert (2013) ACP 9. [15] Gough et al. (2014) EPSL 393. [16] Wise et al. (2012) ACP 12. [17] Nuding et al. (2014) Icarus 243. [18] Primm et al. (2017) GCA 212.  Figure 2. RH and temperature conditions under which (a) CaCl2, (b)  Ca(ClO4)2, and (c) MgCl2 hydrate (solid circles) and dehydrate (open circles). MSL REMS RHg and Tg through sol 1527 (grey circles) and modeled subsurface conditions (10 cm depth) at Ls= 62, 151, 241, and 330. MgCl2 experiments were performed at Earth atmospheric pressures and Mars atmospheric pressures; no difference was observed. 
