•Unusual sterols found in El Junco Lake sediment, including two novel sterols.•Peridinium dinoflagellate was the likely source.•Dinosterol source specificity and abundance make it valuable for paleoclimate research.•Several diols, triols, keto-ols and alkenols can further constrain past environmental change.A variety of lipid biomarkers were identified in sediments from El Junco Lake, Galápagos and their sources investigated for potential use in paleoclimate applications. A series of unusual sterols was also found, including 4α-methylgorgostanol, reported in only four species of dinoflagellates to date. We also tentatively assigned 22,23-methylene-4α-methyl-24-ethylcholest-5-en-3β-ol, the mass spectrum of which matched a sterol found in resting cysts of the dinoflagellate Peridinium umbonatum. In addition, we identified the novel sterol 4α,22,23,24-tetramethyl-5α-cholest-22E-en-3β-ol. Based on the unique sterol distribution, we hypothesize that a dinoflagellate from the genus Peridinium was the primary source of dinosterol and the novel sterols throughout the sediment record. The source specificity and abundance throughout the 3.7 m of recovered sediment make dinosterol an excellent target for hydrogen isotope analysis for use as a paleohydrological proxy in future studies. The abundant C30 and C32 1,ω20-diols and keto-ols, C29 9,10-diol and C29 1,ω9,ω10-triol likely derive from the ferns Azolla microphylla and Cyathea weatherbyana, while sources of the C30 1,ω16-diol and keto-ol, C32 1,ω18-diol and keto-ol, and the C30–C32n-alken-1-ols are likely limited to aquatic microalgae. Due to their source specificity, these diol, triol, keto-ol, and n-alkenol biomarkers present further tools for studying past environmental and climatic change.

We present two new methods for purifying dinosterol (4α,23,24-trimethyl-5α-cholest-22E-en-3β-ol) from sediments for the purpose of hydrogen isotope analysis via gas chromatography–isotope ratio mass spectrometry (GC–IRMS). The first method uses reversed phase-high performance liquid chromatography (RP-HPLC) to purify dinosterol from structurally similar 4α-methyl sterols that co-elute on GC analysis. Dinosterol purified from sedimentary sterol/alcohol fractions using this RP-HPLC method demonstrated an average yield of 80%. A very large isotope effect was observed during RP-HPLC purification, with a 560‰ range in δD value between the first and last 5% of a cholesterol standard, which is four times that during normal phase-HPLC (NP-HPLC) purification. However, we show that dinosterol recombined from 3–4 min of eluent during RP-HPLC purification yields highly reproducible and unbiased isotope values. Due to a larger isotope effect and lower sterol recovery during RP-HPLC, NP-HPLC purification is recommended for samples that do not contain 4α-methyl sterols that co-elute with dinosterol during GC. However, for samples that contain a variety of 4α-methyl sterols, RP-HPLC is more likely to yield baseline resolution of dinosterol. In the second method presented, RP-HPLC purification is preceded by NP-HPLC purification. Using this two step procedure, baseline resolution between dinosterol and all other compounds present was achieved for all samples with an average yield of 60% and, in many cases, dinosterol was purified from all other sedimentary lipids. For samples that contain a variety of 4α-methyl sterols and sitostanol concentration > 2× that of dinosterol, the two step purification method is recommended, as neither NP-HPLC or RP-HPLC alone is likely to yield baseline resolution of dinosterol.Highlights► Two new HPLC methods for purifying dinosterol from other sedimentary alcohols and sterols. ► δD range was 560‰ across a sterol peak during RP-HPLC. ► Average yield 80% for RP-HPLC vs. 60% for two step HPLC method. ► Resolution of dinosterol optimized with a mid-polarity VF-17ms GC column. ► Dinosterol purified via the HPLC methods had reproducible and unbiased δD value.