Serpentinization-Associated Mineral Catalysis of the Protometabolic Formose System

The formose reaction is a plausible prebiotic chemistry, famed for its production of sugars. In this work, we demonstrate that the Cannizzaro process is the dominant process in the formose reaction under many different conditions, thus necessitating a catalyst for the formose reaction under various environmental circumstances. The investigated formose reactions produce primarily organic acids associated with metabolism, a protometabolic system, and yield very little sugar left over. This is due to many of the acids forming from the degradation and Cannizaro reactions of many of the sugars produced during the formose reaction. We also show the heterogeneous Lewis-acid-based catalysis of the formose reaction by mineral systems associated with serpentinization. The minerals that showed catalytic activity include olivine, serpentinite, and calcium, and magnesium minerals including dolomite, calcite, and our Ca/Mg-chemical gardens. In addition, computational studies were performed for the first step of the formose reaction to investigate the reaction of formaldehyde, to either form methanol and formic acid under a Cannizzaro reaction or to react to form glycolaldehyde. Here, we postulate that serpentinization is therefore the startup process necessary to kick off a simple proto metabolic system—the formose protometabolic system.


Formose Reaction or Formose System?
Living systems require complex chemical reactions that are interwoven together [1][2][3]. In order to understand this simply we look to simple experiments in prebiotic chemistry. The formose reaction is a classic experiment from the 1800s predating the field of prebiotic chemistry. Later on, it was interpreted as prebiotic chemistry, analogous to experiments like the Miller-Urey experiment. The formose reaction consists of one reactant-formaldehydereacting under basic conditions to form a complex mixture of sugars [4][5][6][7][8][9][10]. These sugars range from two and three carbon sugars to pentoses, hexoses and larger sugars, branched chain sugars, ketoses, and aldoses. This reaction is prebiotically plausible, being that formaldehyde forms under UV light in planetary atmospheres and could be delivered by meteors [10][11][12][13][14]. Classically, the reaction is catalyzed by calcium ions [4][5][6][7][8][9][10]. The formose reaction is autocatalytic, meaning that the reaction can catalyze itself. This means that its products, such as glycolaldehyde, help to drive the reaction on, further lowering the activation energy cost to produce products. To date, the discussion of the formose reaction has almost exclusively been about sugar production. Ribose is also a product of the formose reaction. Hence, prebiotic chemists have focused on the formose reaction in the hopes of

DFT Calculations
The Minnesota density functional M06-2X [33][34][35] was applied for the present investigation. The basis set used was the standard polarization functions and the diffuse functions augmented valence triple-zeta basis set 6-311+G(d,p) [36]. The Barone-Tomasi polarizable continuum model (PCM) [37] with the standard dielectric constant of water (ε = 78.39) was applied to simulate the solvated environment of an aqueous solution. The force constants were determined analytically in the analysis of harmonic vibrational frequencies for all of the complexes. The DFT method included in the GAUSSIAN 09 program [38] was used for all computations.
All the calculations were performed at the M06-2X/6-311+G(d,p) level. We looked at the gas phase and liquid phase thermodynamic parameters (Gibb's free energy) for the reaction of formaldehyde to either form methanol and formic acid under a Cannizzaro reaction, or to react to form glycolaldehyde in the first step of the formose reaction. Additionally, we computed changes in activation energy barriers in a second set of calculations. These calculations were carried out at the M06-2X/6-311+G(d,p) level with a PCM model (in a water solvent).

Mineral/Chemical Gardens Synthesis and Analysis
Salt pellets were made by compressing CaCl 2 and MgCl 2 (both by Fischer) in a 1:1 ratio by hand. Pellets were placed into a vial containing 1 M sodium meta-silicate (Fischer) solution and allowed to sink to the bottom of the vial. The 1 M sodium meta-silicate solution contained 100 mM Na 2 CO 3 (Fisher). Chemical gardens took about 10 min to grow into the solution at room temperature (25 • C). Chemical gardens were removed from solution, washed with DI water three times to remove external silicates, and allowed to dry overnight. The chemical gardens and minerals, including olivine, serpentinite rocks, and magnetite, were inspected via Raman Spectroscopy and powder X-ray diffractometry (XRD) on a BTX-402 Benchtop XRD and at 785 nm on an Enwave EZI-785-A2 Raman spectroscope. Data were analyzed using Crystal Sleuth Software (RRUFF project 2008 edition).

Formose Reactions Synthesis and Analysis
Formaldehyde solutions were prepared by diluting concentrated formaldehyde with methanol solution, to a final concentration of 1 M formaldehyde containing 0.5 M methanol (Fischer). The methanol was an added stabilizer to the formaldehyde solution, to prevent the formation of paraformaldehyde. The solution was titrated with 1 M NaOH to a final pH of about 12.5. The solution was poured into borosilicate vials. Some solutions had 0.5 g of Life 2023, 13, 1297 4 of 13 powdered mineral or chemical gardens added to them; others had calcium or magnesium chlorides or carbonates added to them as controls, and some vials had no salts or minerals added to them and served as negative controls. Each combination of reagents was tested in triplicate, with all solutions being heated in a hot water bath at 90 • C on a hotplate. Vials were heated for one hour and inspected optically and via time-point sampling with NMR.
After completion, mineral samples and chemical garden samples were filtered out of the solution and inspected via XRD to note any possible changes in structure or ratio.
Solutions were filtered on 0.2 µm pore nylon syringe filters (Thermo Scientific) using 10 mL Hamilton syringes. That is, 0.5 mL of each sample was aliquoted into NMR tubes along with 100 mM 4,4-dimethyl-4-silapentane-1-sulfonic acid (DSS, Sigma-Aldrich) and 10% by volume D 2 O (Sigma-Aldrich). Solutions were characterized using 13 C NMR and 1D proton magnetic resonance measurements with water suppression pulses on a Bruker Neo Advance 600 MHz NMR. Some solutions were spiked with 10 mM of organic acids, including glycolic acid, acetic acid, lactic acid, and formic acid, as described in previous work [9]. Data analyses were run on SpinWorks 4.0 or MNOVA software and SigmaPlot 10.0 software.

Cannizzaro Reaction DFT
Our calculations comparing the formose reaction to the Cannizzaro reaction (Scheme 1) show the predominance of the Cannizaro reaction under many circumstances. Here, we compare the Gibb's free energy values (∆G) of the first step of the formose reaction (the reaction of formaldehyde to glycolaldehyde) to the Cannizaro reaction of formaldehyde (the formation of methanol and formic acid from formaldehyde) ( Table 1).

Formose Reactions Synthesis and Analysis
Formaldehyde solutions were prepared by diluting concentrated formaldehyde with methanol solution, to a final concentration of 1 M formaldehyde containing 0.5 M methanol (Fischer). The methanol was an added stabilizer to the formaldehyde solution, to prevent the formation of paraformaldehyde. The solution was titrated with 1 M NaOH to a final pH of about 12.5. The solution was poured into borosilicate vials. Some solutions had 0.5 g of powdered mineral or chemical gardens added to them; others had calcium or magnesium chlorides or carbonates added to them as controls, and some vials had no salts or minerals added to them and served as negative controls. Each combination of reagents was tested in triplicate, with all solutions being heated in a hot water bath at 90 °C on a hotplate. Vials were heated for one hour and inspected optically and via time-point sampling with NMR.
After completion, mineral samples and chemical garden samples were filtered out of the solution and inspected via XRD to note any possible changes in structure or ratio.
Solutions were filtered on 0.2 µm pore nylon syringe filters (Thermo Scientific) using 10 mL Hamilton syringes. That is, 0.5 mL of each sample was aliquoted into NMR tubes along with 100 mM 4,4-dimethyl-4-silapentane-1-sulfonic acid (DSS, Sigma-Aldrich) and 10% by volume D2O (Sigma-Aldrich). Solutions were characterized using 13 C NMR and 1D proton magnetic resonance measurements with water suppression pulses on a Bruker Neo Advance 600 MHz NMR. Some solutions were spiked with 10 mM of organic acids, including glycolic acid, acetic acid, lactic acid, and formic acid, as described in previous work [9]. Data analyses were run on SpinWorks 4.0 or MNOVA software and SigmaPlot 10.0 software.

Cannizzaro Reaction DFT
Our calculations comparing the formose reaction to the Cannizzaro reaction (Scheme 1) show the predominance of the Cannizaro reaction under many circumstances. Here, we compare the Gibb's free energy values (∆G) of the first step of the formose reaction (the reaction of formaldehyde to glycolaldehyde) to the Cannizaro reaction of formaldehyde (the formation of methanol and formic acid from formaldehyde) ( Table 1).  The above results suggest that products of the Cannizzaro reaction would be more favored than those of the formose reaction. Both reactions prefer the gas phase to the water phase. A lower temperature would benefit both reactions in either the gas or water phase. The Cannizzaro reaction occurs in base conditions. Acidic conditions might benefit the first part of the formose reaction.

Materials Identification
Pellet chemical gardens formed over a period of 10 min, forming from 1:1 CaCl 2 /MgCl 2 pellets in silicate solution ( Figure 1). The pH of the solution was about 12 and it contained sodium carbonate.
Reaction 2: The first step of the formose reaction The above results suggest that products of the Cannizzaro reaction would be more favored than those of the formose reaction. Both reactions prefer the gas phase to the water phase. A lower temperature would benefit both reactions in either the gas or water phase. The Cannizzaro reaction occurs in base conditions. Acidic conditions might benefit the first part of the formose reaction.

Materials Identification
Pellet chemical gardens formed over a period of 10 min, forming from 1:1 CaCl2/MgCl2 pellets in silicate solution ( Figure 1). The pH of the solution was about 12 and it contained sodium carbonate. The chemical gardens were inspected using powder XRD and Raman Spectroscopy to investigate which minerals the crystalline regions were comprised of (Figures 2 and S1). These minerals included dolomite, calcite, aragonite, magnesite, and brucite. The purity of the other minerals was verified spectroscopically as well ( Figures S2-S5). All data were compared to standards on the RRUFF database [39]. Magnetite was not verified this way; it was, however, purchased at 99% purity from Sigma Aldrich. The chemical gardens were inspected using powder XRD and Raman Spectroscopy to investigate which minerals the crystalline regions were comprised of (Figures 2 and S1). These minerals included dolomite, calcite, aragonite, magnesite, and brucite. The purity of the other minerals was verified spectroscopically as well ( Figures S2-S5). All data were compared to standards on the RRUFF database [39]. Magnetite was not verified this way; it was, however, purchased at 99% purity from Sigma Aldrich.

Formose Products
Products of the formose reaction were identified using NMR, including 13 C NMR, spiked-in experiments, and 13 C/ 1 H HSQC ( Figures S7-S12). The formaldehyde solutions we worked with were pure and newly bought stock solutions from Fischer; we also checked the concentrations for accuracy with previously developed methods [40]. These results verify the results of our previous work [9]. Additional controls were run, showing that methanol and paraformaldehyde did not yield similar products when investigated in

Formose Products
Products of the formose reaction were identified using NMR, including 13 C NMR, spiked-in experiments, and 13 C/ 1 H HSQC ( Figures S7-S12). The formaldehyde solutions we worked with were pure and newly bought stock solutions from Fischer; we also checked the concentrations for accuracy with previously developed methods [40]. These results verify the results of our previous work [9]. Additional controls were run, showing that methanol and paraformaldehyde did not yield similar products when investigated in the same fashion that our formose reactions were (data not shown). Additional formose products were identified using glycosyl composition analysis (Table S1, Figures S13-S17). NMR and GCMS analysis indicated several small organic acids associated with metabolism. These included glycolic acid, lactic acid, formic acid, acetic acid, and oxalic acid ( Table 2, Figures S8-S12). This was the first time oxalic acid has been detected in this system; it was not detected in our previous work [9] due to being proton-silent. Glycosyl composition analysis indicated trace amounts of sugars, including glucose, mannose, xylose, and arabinose (Table S1, Figures S13-S17). There were only trace amounts of sugars detected compared to the other products, and sugar yields were extremely low due to sugars degrading in hot alkaline conditions. We verified lactic acid production from sugar breakdown by running a control of an alkaline 0.167 M glucose solution, and we observed similar products in that and other controls (Figures S19 and S20, Table S2).
In order to observe the expanded view of the alkaline formose reaction and its fuller link to metabolism, we note that most products in this system do not end in sugars, and based on the experimental conditions such as basicity and heat Cannizzaro products and tars, they seem inevitable.

Formose System Catalysis with DFT
Our system uses minerals and chemical gardens as heterogenous catalysts. The use of these catalysts and ions in solution as catalysts is an example of Lewis-acid-based catalysis. The reaction of formaldehyde to sugar is followed by a yellowing step; for this yellowing step to occur, sugars must be in solution and then degrade. We conducted multiple time-point experiments to demonstrate the catalytic properties of our minerals and chemical gardens.
In one experiment, we optically observed yellowing of solutions containing added minerals or no added minerals, and we compared average yellowing times ( Figure S18). Solutions containing our chemical garden samples, calcium or magnesium salts, or serpentinite or olivine all reacted much faster than our control group (no minerals/salts added). Additionally, magnetite showed no catalysis ( Figure S18). Interestingly, our chemical garden sample reaction time fell between calcium and magnesium chloride and calcium and magnesium carbonates; it is likely that the garden was leaching ions into the solution.
In addition to faster reaction times in the presence of some of our mineral samples, we looked at product distribution at the 25 min time point to note which solutions had reacted and formed formose products. Here, we observe that the control group (no minerals/salts added) has not formed any formose reaction products, and neither did the magnetite added group (Table S2). At the 25 min time point, we did identify formose products for our chemical garden, olivine, serpentinite, calcium, and magnesium salt groups (Table 3, Figure 3). The control and the magnetite groups showed only Cannizaro products in solution. This was another indicator that some reactions were catalyzed by the added minerals.  Lactic acid, being a sugar breakdown product, is an indicator of the formose reaction and was used to compare formose output between experimental groups. In terms of comparing experimental groups, the highest production of lactic acid was observed in our Ca/Mg mineral-containing samples (around 0.5 int. %), whereas serpentinite and olivine showed the second-highest amount lactic acid (0.36 and 0.31 int. %, respectively). Magnetite and the no mineral group showed no lactic acid (Table 3).
In order to further explain the observed catalytic process, we propose a mechanism for the formose reaction and use our DFT calculations to model what happens at each step of this mechanism (Scheme 2 and Table 4). Our mechanism addresses the slow part of the formose process-the formation of glycolaldehyde from formaldehyde. Lactic acid, being a sugar breakdown product, is an indicator of the formose reaction and was used to compare formose output between experimental groups. In terms of comparing experimental groups, the highest production of lactic acid was observed in our Ca/Mg mineral-containing samples (around 0.5 int. %), whereas serpentinite and olivine showed the second-highest amount lactic acid (0.36 and 0.31 int. %, respectively). Magnetite and the no mineral group showed no lactic acid (Table 3). In order to further explain the observed catalytic process, we propose a mechanism for the formose reaction and use our DFT calculations to model what happens at each step of this mechanism (Scheme 2 and Table 4). Our mechanism addresses the slow part of the formose process-the formation of glycolaldehyde from formaldehyde.

Mineral System Evolution
In terms of understanding how our mineral systems changed during the reactions, we analyzed our minerals with powder XRD after our hydrothermal reactions were carried out. Our magnetite, olivine and serpentinite samples were not changed during our hydrothermal reactions (data not shown). The composition of the chemical garden sample did change, however (Figure 4). The garden composition changed due to hydrothermal alteration, from a composition rich in magnesium and calcium carbonates and hydroxides

Mineral System Evolution
In terms of understanding how our mineral systems changed during the reactions, we analyzed our minerals with powder XRD after our hydrothermal reactions were carried out. Our magnetite, olivine and serpentinite samples were not changed during our hydrothermal reactions (data not shown). The composition of the chemical garden sample did change, however (Figure 4). The garden composition changed due to hydrothermal

Mineral System Evolution
In terms of understanding how our mineral systems changed during the reactions, we analyzed our minerals with powder XRD after our hydrothermal reactions were carried out. Our magnetite, olivine and serpentinite samples were not changed during our hydrothermal reactions (data not shown). The composition of the chemical garden sample did change, however (Figure 4). The garden composition changed due to hydrothermal

Mineral System Evolution
In terms of understanding how our mineral systems changed during the reactions, we analyzed our minerals with powder XRD after our hydrothermal reactions were carried out. Our magnetite, olivine and serpentinite samples were not changed during our hydrothermal reactions (data not shown). The composition of the chemical garden sample did change, however (Figure 4). The garden composition changed due to hydrothermal

Mineral System Evolution
In terms of understanding how our mineral systems changed during the reactions, we analyzed our minerals with powder XRD after our hydrothermal reactions were carried out. Our magnetite, olivine and serpentinite samples were not changed during our hydrothermal reactions (data not shown). The composition of the chemical garden sample did change, however (Figure 4). The garden composition changed due to hydrothermal alteration, from a composition rich in magnesium and calcium carbonates and hydroxides

Mineral System Evolution
In terms of understanding how our mineral systems changed during the reactions, we analyzed our minerals with powder XRD after our hydrothermal reactions were carried out. Our magnetite, olivine and serpentinite samples were not changed during our hydrothermal reactions (data not shown). The composition of the chemical garden sample did change, however (Figure 4). The garden composition changed due to hydrothermal

Mineral System Evolution
In terms of understanding how our mineral systems changed during the reactions, we analyzed our minerals with powder XRD after our hydrothermal reactions were carried out. Our magnetite, olivine and serpentinite samples were not changed during our hydrothermal reactions (data not shown). The composition of the chemical garden sample did change, however (Figure 4). The garden composition changed due to hydrothermal

Mineral System Evolution
In terms of understanding how our mineral systems changed during the reactions, we analyzed our minerals with powder XRD after our hydrothermal reactions were carried out. Our magnetite, olivine and serpentinite samples were not changed during our hydrothermal reactions (data not shown). The composition of the chemical garden sample did change, however (Figure 4). The garden composition changed due to hydrothermal alteration, from a composition rich in magnesium and calcium carbonates and hydroxides

Mineral System Evolution
In terms of understanding how our mineral systems changed during the reactions, we analyzed our minerals with powder XRD after our hydrothermal reactions were carried out. Our magnetite, olivine and serpentinite samples were not changed during our hydrothermal reactions (data not shown). The composition of the chemical garden sample did change, however (Figure 4). The garden composition changed due to hydrothermal alteration, from a composition rich in magnesium and calcium carbonates and hydroxides

Mineral System Evolution
In terms of understanding how our mineral systems changed during the reactions, we analyzed our minerals with powder XRD after our hydrothermal reactions were carried out. Our magnetite, olivine and serpentinite samples were not changed during our hydrothermal reactions (data not shown). The composition of the chemical garden sample did change, however (Figure 4). The garden composition changed due to hydrothermal alteration, from a composition rich in magnesium and calcium carbonates and hydroxides indicates a mechanism for the catalysis of the formose process by calcium and magnesium ions/minerals.
All the calculations were carried out at the M06-2X/6-311+G(d,p) level with a PCM model (solvent = water). From step 4 onwards (Scheme 2), Ca 2+ and Mg 2+ have little effects on the activation energies. For the entire reactions (from formaldehyde to glycolaldehyde), the results suggest that the rate-determining step should be Part 1 (steps 1-3). The steps in Part 2 (steps 4-10) have energy barriers that reveal that those reactions can be observed at room temperature. Glycolaldehyde might be helpful to condense formaldehyde into more glycolaldehyde. However, calcium and magnesium do not show many effects on accelerating the related reactions (as in steps 4-10). We did not show the null results for steps 4-10; we only showed the steps that had a change in activation energy, i.e., steps 1-3.

Mineral System Evolution
In terms of understanding how our mineral systems changed during the reactions, we analyzed our minerals with powder XRD after our hydrothermal reactions were carried out. Our magnetite, olivine and serpentinite samples were not changed during our hydrothermal reactions (data not shown). The composition of the chemical garden sample did change, however ( Figure 4). The garden composition changed due to hydrothermal alteration, from a composition rich in magnesium and calcium carbonates and hydroxides to a composition of primarily calcium carbonates such as calcite and aragonite. We replicated this change using hot water (instead of formaldehyde solution) ( Figure S6), showing that the garden composition change was due to hydrothermal alteration and not the formose reaction. to a composition of primarily calcium carbonates such as calcite and aragonite. We replicated this change using hot water (instead of formaldehyde solution) ( Figure S6), showing that the garden composition change was due to hydrothermal alteration and not the formose reaction. . Powder X-ray diffraction pattern for a Ca/Mg-Based chemical garden (top) before being exposed to the formose reaction. Powder X-ray diffraction pattern for a Ca/Mg-Based chemical garden (bottom) after being exposed to the formose reaction. The garden is comprised of microcrystalline B = brucite, C = calcite, A = aragonite, M = magnesite, and D = dolomite. The garden is also comprised of amorphous calcium and magnesium, carbonates, hydroxides, and silica. Base line calibrated with Crystal Sleuth software.

Main Points Summarized
In this work, we demonstrate that the Cannizaro process is the dominant process under many different conditions, thus necessitating a catalyst for the formose reaction under many different environmental circumstances. Our formose reactions produce primarily organic acids associated with metabolism and a protometabolic system, and have very little sugar left over. This is due to many of our acids forming from the degradation and Cannizaro reactions of many of the sugars produced during the formose reaction. . Powder X-ray diffraction pattern for a Ca/Mg-Based chemical garden (top) before being exposed to the formose reaction. Powder X-ray diffraction pattern for a Ca/Mg-Based chemical garden (bottom) after being exposed to the formose reaction. The garden is comprised of microcrystalline B = brucite, C = calcite, A = aragonite, M = magnesite, and D = dolomite. The garden is also comprised of amorphous calcium and magnesium, carbonates, hydroxides, and silica. Base line calibrated with Crystal Sleuth software.

Main Points Summarized
In this work, we demonstrate that the Cannizaro process is the dominant process under many different conditions, thus necessitating a catalyst for the formose reaction under many different environmental circumstances. Our formose reactions produce primarily organic acids associated with metabolism and a protometabolic system, and have very little sugar left over. This is due to many of our acids forming from the degradation and Cannizaro reactions of many of the sugars produced during the formose reaction. Hence, our study mainly focuses on organic acids. We also show heterogeneous Lewis-acid-based catalysis of the formose reaction by mineral systems associated with serpentinization. The minerals that showed catalytic activity include olivine, serpentinite, and calcium, and magnesium minerals including dolomite, calcite, and our Ca/Mg-chemical gardens. This catalysis is due to the first three steps of the reaction, where formaldehyde reacts to form glycolaldehyde in the presence of calcium or magnesium.

On the Potential Ubiquity of the Formose Proto-Metabolic Systems Due to Serpentinization
The origins and emergence of life need not be constrained to one environment or one system; any mineral-rich system could do, especially a hydrothermal system [41][42][43][44]. Another environment to consider would include potentially space-based environments that have a similar mechanisms of catalysis based on magnesium minerals [45]. The products in these systems may not necessarily be analogous to classical Butlerov formose-yielding sugars [9,44,45], but they are significant, nonetheless.
The serpentinization of dry land systems can lead to pools or ponds of water in those systems becoming more alkaline, not only facilitating the pH needed for the formose reaction, but also the formation of Ca/Mg carbonate minerals needed for catalysis, as alkaline conditions absorb more CO 2 (Scheme S1). Obviously, deep-sea systems could also exhibit this phenomenon, being that they can form atop serpentinizing crust and form mineral chimneys analogous to our chemical gardens (Scheme S2). These systems are rich with energy and produce catalysts for the formose reaction, and while conditions are not favorable for sugar stability, they do yield a protometabolic system of very simple organic acids, including lactic acid, formic acid, acetic acid, glycolic acid, and oxalic acid. It could be that serpentinization is how terrestrial planets and ice moons "breathe" life into protometabolic systems, the simplest example being the formose protometabolic system.

Supplementary Materials:
The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/life13061297/s1, Figure S1: Raman Spectrum of Ca/Mg chemical garden. Spectrum taken with a 780 nm laser. Verified against RRUFF mineral database; Figure S2: Raman Spectrum of olivine sample. Spectrum taken with a 780 nm laser. Verified against RRUFF mineral database Figure S3: Raman spectrum of serpentinite. Spectrum taken with a 780 nm laser. Verified against RRUFF mineral database; Figure S4: Powder XRD diffraction pattern for olivine. 5-55-degree angle used. Verified against RRUFF mineral database; Figure S5: Powder XRD diffraction pattern for serpentine. 5-55-degree angle used. Verified against RRUFF mineral database; Figure S6: Hydrothermal alteration of the chemical garden. Hydrothermal alteration of the chemical garden occurs during the heating of the water for the formose process. During this process, the chemical garden transitions from a composition of calcite, magnesite, aragonite, brucite, and dolomite to a composition of mostly calcite, with some aragonite and Mg left; Figure S7: Concentrated Formose reaction. Chemical Garden facilitated reaction, dried overnight in the fume hood. Proton NMR with customized water suppression (noesygppr1d) technique on the Brucker Neo 600 with D2O Salt Locking. No peak at 3.2 ppm for methanol due to evaporative process; Figure S8: 13C NMR of a Ca/Mg-Based chemical garden facilitated formose reaction. Spectrum shows detection of glycolic acid, G, acetic acid, A, and lactic acid, L. Spectrum also indicates the presence of oxalic acid, Ox, which has yet to be detected in such studies due to it being proton silent; Figure S9: Formose solution with no spiked in products; Figure S10: Spiked in-experiment. Formose solution was spiked with glycolic acid, G, acetic acid, A, and lactic acid, L; Figure S11: Spiked in-experiment. Formose solution was spiked with glycolic acid, G, acetic acid, A, and lactic acid, L (the blue spectrum). This is overlayed onto a