Banded Iron Formations (BIFs) are ancient marine chemical sediments that contain various Fe-bearing minerals such as hematite (Fe2O3), magnetite (Fe3O4), siderite (FeCO3) and a variety of FeII-/FeIII-silicates. The prevailing opinion is that primary Fe(III) (oxyhydr)oxides, such as ferrihydrite (simplified formula of Fe(OH)3), were precipitated from the ocean’s photic zone by marine plankton, and a fraction of these minerals was subsequently transformed into secondary magnetite and siderite by dissimilatory Fe(III)-reducing bacteria (DIRB). However, aside from broad estimates, it is currently unknown what fraction of the primary Fe(III) minerals was sedimented to the seafloor where it was eventually lithified, and what fraction was reduced by DIRB in the water column, thus forming a microbial Fe cycle in the water column. To test this, we conducted Fe cycling experiments with marine phototrophic Fe(II)-oxidizing bacteria and DIRB under conditions mimicking the Precambrian ocean water column with elevated Fe(II) and Si concentrations. We followed secondary mineral formation over three consecutive redox cycles (oxidation followed by reduction) over a time interval of up to 58 days to determine which mineral phases would ultimately have settled as BIF forming sediments. We used wet geochemical methods to follow Fe speciation, measured dissolved silica and volatile fatty acid (VFA) concentrations, determined cell-mineral associations using fluorescence and electron microscopy, and characterized the mineralogy of the precipitates using 57Fe-Moessbauer spectroscopy and X-ray diffraction (XRD). Our results showed that both the absence of silica and an increasing number of Fe cycles favored the formation of more crystalline minerals, such as goethite (α-FeOOH). However, in the presence of high concentrations of monomeric silica, as suggested for ancient oceans (2.2 mM), only short-range ordered (SRO) Fe(III) minerals such as ferrihydrite were observed. These did not transform into the more thermodynamically stable goethite during repeated Fe cycling. Interestingly, no magnetite formed in any of the setups. Instead, increasing Si concentrations favored the formation of increasing quantities of Fe(II) minerals. Microscopy revealed a tight association between microbial biomass and minerals formed. Dissolved silica analysis showed the removal of Si from solution congruent with Fe(II) oxidation and a release of Si during Fe(III) reduction. Together, these results suggest an important role of co-precipitated biomass as well as silica for secondary mineral formation by either constraining crystal growth and/or inhibiting Fe(II)-induced mineral transformation. Overall, our results imply that microbial Fe cycling during settling of primary ferrihydrite through the photic zone in a Precambrian ocean would have resulted in the partial transformation of ferrihydrite into secondary Fe(II) mineral phases in the water column. This would have resulted in the accumulation of mixtures of a ferrihydrite-silica composite and Fe(II) minerals in the initial BIF forming sediments.