Crystallization experiment
In order to simulate the high-pressure conditions at the seafloor, three laboratory experiments involving aragonite precipitation were conducted.  Two experiments were conducted using a high-pressure apparatus (Figure 1). The control experiment was performed at atmospheric pressure and 8°C in a container immersed in a constant temperature bath. The experimental fluid (growth medium) was prepared by mixing artificial seawater (ASW) and small aliquots of Na2CO3. The addition of Na2CO3 allowed the saturation state of the fluid with respect to aragonite to be increased, in order to promote nucleation and crystallization on carbonate minerals. Artificial seawater was prepared by dissolving 32 g of Instant Ocean salt (Instant Ocean Spectrum Brands) mix in 1 kg of distilled water and filtering it through a 0.2 µm Nalgene filter. A small amount of 0.1M Na2CO3 was then introduced into the continuously stirred artificial seawater using a syringe pump set at a rate of ~1 ml/min, to reach a final volume ratio of Na2CO3 to artificial seawater of ~1:13. During the addition of Na2CO3, the solution was in contact with the atmosphere. This procedure homogenized the fluid and also precluded aragonite nucleation prior to the beginning of the experiment. The absence of crystal nucleation was confirmed visually by monitoring the solution pH, which remained stable for the 1-2 hours needed for the experiment setup. The observation that aragonite crystallization did not start prior to the beginning of the experiment is consistent with previously published data, which show that 6 hours elapsed before onset of crystallization in the solution similar to the one we used (7.4 mmol of Na2CO3 :1 liter of seawater) (Pytkowicz, 1965).

Petrographic glass slides frosted with silicon carbide (SiC) abrasive paper were used to promote aragonite crystallization along scratches. For each experiment, one slide was placed into the high-pressure vessel prior to the transfer of the experimental fluid. The experimental fluid was then placed into the vessel after it was precooled to 8°C. After that, the air space above the solution (<200 cm3) was purged with nitrogen three times before the pressure was increased to a desired value (110 or 345 bars). The experimental fluid was sampled daily using a valve on the pressure vessel (Figure 1) and pH was measured to estimate time of nucleation and crystallization of CaCO3. For pH measurements, electrode calibration and measurements were conducted at temperatures similar to those of the experiment. NBS buffer solutions with pH of 4, 7, and 10 were used for electrode calibration. The pH initially decreased due to carbonate precipitation and reached near-constant values less than 170 hrs from the beginning of the experiments, suggesting that most of the CaCO3 precipitated in the first few days of each experiment. At the end of each run, the experimental products were extracted from the pressure vessel. Crystals (<10 µm in size) were rinsed with distilled water and dried, whereas the experimental fluid was filtered and refrigerated for subsequent elemental analyses. The alkalinity of the final fluid was measured the day after the end of each experiment. Experimental parameters are presented in Table -1.