August 2017 Issue Table of Contents
The boron isotope composition of calcium carbonate shells of marine organisms has the unique potential to record surface ocean pH, allowing the calculation of atmospheric pCO2 due to the established relationship between pH and the partial pressure of (atmospheric) CO2 (pCO2). This “paleo-pH meter” allows scientists to produce a record of the natural fluctuations of atmospheric pCO2 over geologic time, which will help us better understand the impacts of the recent anthropogenic addition of CO2 to Earth’s atmosphere. Towards this end, a tremendous effort to understand the systematics of boron uptake in marine carbonates is underway. Here, we review the potential of boron isotopes to constrain ocean pH and, thus, atmospheric pCO2.
Keywords: pH proxy, boron isotopes, pCO2, foraminifera, brachiopods, paleoclimate
The isotope ratio of light elements, such as oxygen and carbon, have proved extremely valuable for studies related to biological and geochemical processes due to the large mass difference between the isotopes of those elements. These light isotope systems are widely used in paleoclimate research, because climate processes result in a large range in isotope compositions of geologic and biologic materials. The past few decades have seen an impressive increase in the application of other stable light isotopes, each with its own potential for illuminating specific processes. Early studies of boron isotopes in marine carbonates had the goal of exploiting the large percent mass difference between 11B and 10B, the two stable isotopes of boron, similar to the way carbon and oxygen isotopes have been applied. However, in contrast to those other stable isotope systems, the mass-dependent boron isotope fractionation between seawater and calcium carbonate shells is not the dominant controlling factor. Instead, the major control on boron isotope fractionation is related to the effect of pH on the aqueous speciation of boron and to an observed large offset in the isotopic composition between the two primary aqueous species. Incorporation of only one of these species into marine carbonates leads to a pH control on the isotopic composition of the carbonate minerals, and thus, the ability to estimate ocean pH in the past using well-preserved fossil shells. With an estimate of past surface ocean pH, and knowledge of one additional aqueous carbonate system variable (e.g. alkalinity, total dissolved inorganic carbon), then atmospheric pCO2 can be calculated due to the strong coupling of the surface ocean and atmosphere. Due to this unique potential of the boron proxy for estimating pCO2, a tremendous effort has focused on a more fundamental understanding of boron speciation and isotope fractionation and on boron incorporation into carbonates in order to improve the application of this geochemical tool. Equally important are the analytical developments, including international comparisons between laboratories (Foster et al. 2013) as well as single laboratory comparisons between the two most used measurement methods, those of negative thermal ionization mass spectrometry (NTIMS) and of multi-collector inductively coupled plasma mass spectrometry (MC–ICP–MS) (Farmer et al. 2016). Progress in these areas is bringing us closer to the routine and widespread application of boron isotopes in geochemical studies. Here, we review the basis for the boron pH proxy and a sample of recent papers that illustrate the unique potential of boron as a pH proxy for seawater, and thus, for atmospheric pCO2.
THE BORON pH PROXY
The aqueous speciation of boron is controlled by pH, with boron in trigonal (3-fold) coordination in boric acid [B(OH)3] dominating at low pH and boron in tetrahedral (4-fold) coordination in borate [B(OH)4−] dominating at high pH (Fig. 1). Boric acid has a preference for the heavier isotope, 11B, whereas borate has a preference for the lighter isotope, 10B. This preference, or fractionation factor, predicts an isotope ratio of each boron species, which in turn reflects the proportions of boron species in an aqueous fluid having 3-fold and 4-fold coordination (Fig. 1). Incorporation of only borate in carbonate shells is the fundamental mechanism that allows the boron isotope composition of marine carbonate fossils to record the pH of ancient oceans.
Boron Incorporation into Carbonates
The large offset between the boron isotope composition of seawater and marine carbonates was first observed in studies of modern biogenic and abiogenic minerals. The isotope composition of these minerals was significantly offset towards lighter boron isotope values: it was much closer to the boron isotope composition of the B(OH)4− aqueous species than to the bulk composition of seawater. This led to the interpretation that only the borate ion is incorporated into the carbonate structure (Hemming and Hanson 1992). More recently, the fractionation factor has been empirically determined to be 27.2 ± 0.6‰, based on dissociation constants for single isotope borate buffer solutions (Klochko et al. 2006). The d11B value is the per mil deviation from the isotopic composition of the widely used NIST SRM 951 boric acid isotopic standard, and the empirically derived d11B versus pH curves for seawater based on single species foraminifera grown in controlled pH conditions can be significantly different than the measured fractionation factor would predict (see Foster and Rae 2016). This means that empirical data are needed for each species in order to calculate ocean pH.
Laboratory investigations of boron isotopes have used inorganic carbonates synthesized under controlled pH conditions to evaluate boron incorporation (Xiao et al. 2006; Mavromatis et al. 2015; Noireaux et al. 2015; Uchikawa et al. 2015; Kaczmarek et al. 2016). Without exception, these studies show a strong dependence of the boron isotope composition on pH and a significant offset from the parent fluid. Hemming and Hanson (1992) theorized that the charged borate ion is attracted to the growing mineral surface, but is ultimately incorporated in 3-fold coordination. Based on nuclear magnetic resonance (NMR) data, calcite has both 3-fold and 4-fold coordinated boron, which has been interpreted either to represent an incomplete change in the borate ion at the mineral surface before incorporation (Sen et al. 1994) or that significant boric acid is assimilated (Klochko et al. 2009; Rollion-Bard et al. 2011; Cusack et al. 2015; Noireaux et al. 2015). However, even small amounts of boric acid would have a measurable impact on the d11B of the resulting calcite, and this fact, combined with the expectation of greater accidental incorporation in rapidly deposited calcite, is at odds with the usual observed trend of greater deviation from the borate value with lower pH (Foster and Rae 2016). Using synchrotron X-ray nanoscale techniques on a single foraminifera shell, Branson et al. (2015) showed no measurable 4-coordinated boron in foraminiferal calcite. Therefore, the only reasonable explanation to account for the observed isotope offset from the parent fluid requires a crystallographic control on modification of that species so that it can fit in the crystal lattice.
Culture experiments of foraminifera to test the boron proxy have yielded empirical data of boron isotopes in carbonates formed over a range of pH values. This has been summarized in detail in a recent review of the potential use of boron isotopes in foraminifera as a paleo-pH proxy (Foster and Rae 2016). The bottom line is that, with few exceptions, biogenic calcite does not give the isotope value of borate for the pH of the seawater in which it formed, yet it does have a systematic and predictable offset from the borate value. Determining this offset requires empirical calibrations to estimate pH. The reasons for this may be related to the control the organism has on the secretion of its shell (although this is seen in inorganic experiments as well), and it appears to be more pronounced for lower pH solutions. Perhaps organisms exert more of a control to raise pH when they are in lower pH environments, thus influencing this offset from the actual fractionation factor value. Another potential factor is that organisms often build their shells from amorphous calcium carbonate that subsequently crystallizes to calcite or aragonite. Could secondary processes related to the conversion to the crystalline phase partly control boron behavior? For example, it is easier to see how borate, without change in coordination, might be incorporated in amorphous calcium carbonate, and then remain in the structure even after conversion to the carbonate mineral. Henehan et al. (2016) have compiled new and published empirical data from planktic foraminifera using a graph of expected borate isotopic composition based on oceanographic conditions and the fractionation factor of Klochko et al. (2006) versus the boron isotope compositions of the foraminifera (Fig. 2). This clever plot assumes that all of the incorporated boron is borate. However, if small amounts of boric acid were systematically incorporated then this could be accounted for with the empirical equations. This compilation shows that some species plot at heavier-than-predicted compositions, consistent with the site of incorporation having a higher-than-seawater pH, few plot on the one to one line, and some plot at lower d11B than predicted from measured oceanographic conditions. Possible reasons for these differences are discussed in Foster and Rae (2016).
Foraminifera culture experiments that monitored for changes in dissolved inorganic carbon show that the carbon speciation also changes depending on the changing conditions (such as pH) of the experiments (Allen et al. 2011). In a study that controlled for pH in some experiments and [CO2] in others, and in each case holding the other parameter constant, Howes et al. (2017) demonstrated that whereas d11B incorporated in carbonates is controlled by pH, [B] is controlled by [CO32−]. This makes sense in that the isotope composition is controlled by borate incorporation (regardless of quantity), whereas the borate ion competes for the crystallographic site that is occupied by CO32−. It is clear that there are some rich avenues for research to further our understanding of the boron pH proxy, and that boron isotopes also have the potential to be used to better understand the controlling factors in biomineralization. Importantly, without exception, every published empirical study shows a large offset in the boron isotopic signature between the carbonate mineral and the solution from which it grew, and this offset varies systematically with the pH of the solution. Secondary factors exist, requiring the need for calibrations of each biological carbonate used in paleo-pH studies, but these do not invalidate the proxy.
Application to Testing pH Change in the Marine Carbonate Record
The high-resolution studies of glacial to interglacial changes in CO2 determined from bubbles trapped in Greenland and Antarctic ice cores provide an excellent test of the potential for boron isotopes to accurately record atmospheric pCO2. Several studies that examined the d11B of planktic forams from ocean cores are remarkably consistent with the ice-core records of CO2 (Fig. 3), giving confidence that boron isotopes can be used in deeper geologic time where ice-core records are not available.
Taking the Boron pH Proxy Back in Time
To apply the boron pH proxy through geologic time requires three things. First, that the isotope composition of seawater be estimated. Second, that there be a calibration of the carbonate being measured. Third. that the carbonate has not been altered by diagenesis. All of these requirements become increasingly difficult to meet in older samples, particularly for whole-rock samples. Fortunately, robust methods to screen for diagenetic alteration have already been well-established (Edgar et al. 2015; Stewart et al. 2015). As we will show, even though neither the seawater composition nor the detailed calibrations of biological carbonates are known throughout the Phanerozoic, careful studies of well-preserved fossils are placing limits on the system, and so are contributing to a better understanding on what controls secular variability as well as the magnitude of carbon cycle perturbations.
The Boron Isotope Composition of Ancient Oceans
The boron isotope composition of the fluid, in this case seawater, must be known or assumed in order to calculate pH from the d11B of carbonates. One major feature of seawater that must be accounted for in models of secular change is that the d11B of seawater is remarkably heavier than the known primary input, i.e. river water, wich has an estimated average d11B of 10‰ (Lemarchand et al. 2002). Future studies should also consider the importance of the subterranean groundwater estuary, which is equal in volume to rivers and is a diurnal reactor bed of geochemical processing. The greatest recognized sink is mid-ocean ridge (MOR) hydrothermal fluid–rock interactions (Simon et al. 2006). This, and all known fluxes out of the system, involves removal of isotopically light boron, which accounts for the major enrichment to 39.6‰ in present-day seawater. To better understand the secular variability of d11B in seawater, researchers have turned to process-based flux models (Lemarchand et al. 2002; Joachimski et al. 2005; Simon et al. 2006). From the accumulated boron isotope data, as well as the modeling studies, it is clear that a primary control on d11B of seawater is the removal of light boron. It is further clear that this control is primarily due to the reactive nature of the charged isotopically light borate ion that is more likely to be sorbed or incorporated through any of the alteration and precipitation processes that are removing boron. As a higher resolution record of secular variability emerges, factors that might control short-term changes in the d11B of seawater will come into focus.
Information on the secular variation of seawater in the Cenozoic is now emerging. A careful study using carbon isotopes to reconstruct the pH gradient demonstrates that d11B of seawater has changed by ~3‰ over the past 50 million years (Greenop et al. 2017). Interestingly, instead of a steady increase, it appears that seawater d11B shows a major jump between the Middle to Late Miocene of about 2‰ per million years (Fig. 4). This is a major consideration for using d11B to consider the magnitude of pH change, as the entire range for the glacial/interglacial cycles is about 0.6‰. Still, considering short (sub-million year) timescales, changes in the B/Ca and d11B can place bounds on pH and the changes in dissolved inorganic carbon that accompany perturbations of the carbon system. For example, Penman et al. (2014) used B/Ca and d11B in planktonic foraminifera to study rapid change across the Paleocene–Eocene Thermal Maximum, an interval that has long been known to have a major addition of light carbon associated with significant global warming. The boron results require there to be a rapid and sustained input of carbon to the system and represent a first estimate of the pH change associated with this climate event. Foster and Rae (2016) used a Monte Carlo approach to estimate the errors on the pH from this dataset and found that the actual d11B of seawater, and the pH prior to the perturbation, exert only a small influence on the estimates of the pH change. Importantly, the Paleocene–Eocene Thermal Maximum is considered to be a good analogue for the current rapid increase in CO2 to the Earth’s atmosphere resulting from anthropogenic emissions.
Foraminifera can be planktic or benthic and they tend to occupy narrow depth ranges. Decreasing temperature and increasing hydrostatic pressure though the water column allows more CO2 to be stored with depth and, therefore, the pH and alkalinity shows a gradient. There is also a coherent and known relationship between oxygen isotopes and temperature. Anagnostou et al. (2016) used oxygen isotopes in foraminifera from ocean sediment drill cores to estimate the habitat depth and were able to apply the gradient in the measured d11B to estimate the d11B of seawater. Then, applying pH calibrations based on empirical data from foraminifera, they were able to reconstruct pCO2 through the Eocene warm interval into the Neogene cooling. As with the Penman et al. (2014) study, Anagnostou et al. (2016) showed a clear change in the pH of the oceans over the studied stratigraphic depth interval despite uncertainties in the calibration and the secular variation of d11B of seawater. This work shows >1,000 ppm CO2 for the Early Eocene Climate Maximum at about 50 Ma, declining to less than 700 ppm by 40 Ma when glaciation is known to have started in Antarctica. This decline is coincident with a marked increase in d18O, which suggests that the temperatures declined by 5 °C. Anagnostou et al. (2016) used these relationships to consider the equilibrium climate sensitivity (temperature change in degrees centigrade for a doubling of atmospheric CO2) and found that it is comparable to that of today.
Taking the pH proxy to the pre-Cenozoic is even more of a challenge. The boron isotope system is highly susceptible to diagenetic resetting, aragonite fossils are rarely preserved, and many of the fossils with calcite shells that appear to be good choices for preserving the original boron isotope signature are extinct, precluding the establishment of empirical pH–d11B curves. Of the published pre-Cenozoic studies, brachiopods and rugose corals appear to have good potential as archives with relatively high boron concentrations. Trilobite fossils are typically calcite with very good exoskeleton preservation and could be good archives of boron isotopes, but no published data are available. Rugose corals and trilobites are extinct and have no modern representatives, but reliability could be tested by comparing fossils from different localities and different fossils from one horizon. Penman et al. (2013) examined a variety of modern brachiopods from known pH and found that there appears to be a species control on borate incorporation. This study also showed that B isotopes show variability across the brachiopod shell, as do C and O isotopes, which may also be a function of vital effects. Whereas these differences result in several per mil offsets from the borate curve, as well as differences from each other, there is a fairly solid 23‰ offset from seawater. Most of the measurements plot below the Klochko et al. (2006) borate curve, similar to non-symbiont-bearing benthic foraminifera as discussed in Foster and Rae (2016). To account for the differences among species, it is important to analyze the same species as well as different species from the same horizons to evaluate how different they might be. At this stage, the small differences among species are similar to the variation seen within similar time slices from the stratigraphic record. Future studies could target intervals of known change to see how boron isotopes respond, and it is expected that patterns will emerge that will offer important insights into the Earth system response to changing conditions. Emerging data from the Carboniferous–Permian glacial interval shows a major shift to values similar to the Neogene (Legett et al. 2016) (Fig. 5). Whereas the ancient record is currently sparse, the existing data show much greater variability prior to the major proliferation of planktic calcifiers during the Mesozoic. The rise and diversification of calcifying organisms shifted carbonate deposition from the continental shelf to throughout the oceans. This innovation may have changed the regulation of ocean chemistry so that extremes in pH are now not allowed (Ridgwell 2005). The Joachimski et al. (2005) model for d11B in the Paleozoic is an impressive fit to the compiled data (Fig. 5). As a higher resolution record of d11B and elemental changes in seawater composition emerges, we expect that the nature of boron removal from seawater will offer a unique perspective into controls on seawater chemistry, as well as to changing atmospheric pCO2.
Boron isotopes in marine carbonates are already proving extremely valuable for paleoclimate studies. The prospects now look bright for testing models of the secular variability of the ocean and atmosphere and having a possible tool for studying the controls on biomineralization. Boron isotope composition of carbonates is demonstrably a measure of the pH of the fluids from which the carbonates formed. As with any proxy, calibrations are required. With a direct measure of the fractionation factor, and empirical data from culture experiments, we can now consider what are the controls on d11B and B/Ca in biogenic carbonates. Clearly, taking this further back in time involves making assumptions about how extinct species controlled their local environment, and, perhaps even more difficult, the magnitude of secular variability. Application of the boron pH proxy to the Precambrian, before the advent of multicellular organisms, will require even more assumptions. Any study of Precambrian rocks must consider the control that diagenesis imposes on the boron system. Typically, diagenesis would produce scatter such that multiple analyses from the same horizons, coupled with multiple geologic sections of the same age from widely separated parts of the globe, would be needed to test the reliability of this proxy in Precambrian records.
Don Penman and an anonymous reviewer are thanked for their thoughtful review of the original manuscript. Ed Grew is thanked for a careful edit of the revision which has improved the clarity. Katie Wooton helped with drafting figures. This work is supported in part by NSF-EAR1324725 grant to Hemming and Rasbury. We acknowledge Cara Thompson’s unpublished data from the Ordovician as well as Shelbie Legett and Don Penman’s unpublished data from the Late Paleozoic that are used in Figure 5.
Allen KA and 5 coauthors (2011) Controls on boron incorporation in cultured tests of the planktic foraminifer Orbulina universa. Earth and Planetary Science Letters 309: 291-301
Anagnostou E and 8 coauthors (2016) Changing atmospheric CO2 concentration was the primary driver of early Cenozoic climate. Nature 533: 380-384
Branson O and 7 coauthors (2015) The coordination and distribution of B in foraminiferal calcite. Earth and Planetary Science Letters 416: 67-72
Cusack M, Kamenos NA, Rollion-Bard C, Tricot G (2015) Red coralline algae assessed as marine pH proxies using 11B MAS NMR. Scientific Reports 5, doi: 10.1038/srep08175
Edgar KM, Anagnostou E, Pearson PN, Foster GL (2015) Assessing the impact of diagenesis on δ11B, δ13C, δ18O, Sr/Ca and B/Ca values in fossil planktic foraminiferal calcite. Geochimica et Cosmochimica Acta 166: 189-209
Farmer JR, Hönisch B, Uchikawa J (2016) Single laboratory comparison of MC-ICP-MS and N-TIMS boron isotope analyses in marine carbonates. Chemical Geology 447: 173-182
Foster GL and 9 coauthors (2013) Interlaboratory comparison of boron isotope analyses of boric acid, seawater and marine CaCO3 by MC-ICPMS and NTIMS. Chemical Geology 358: 1-14
Foster GL, Pogge von Strandmann PAE, Rae JWB (2010) Boron and magnesium isotopic composition of seawater. Geochemistry, Geophysics, Geosystems 11, doi: 10.1029/2010GC003201
Foster GL, Rae JWB (2016) Reconstructing ocean pH with boron isotopes in foraminifera. Annual Review of Earth and Planetary Sciences 44: 207-237
Greenop R and 8 coauthors (2017) A record of Neogene seawater δ11B reconstructed from paired δ11B analyses on benthic and planktic foraminifera. Climate of the Past 13: 149-170
Hemming NG, Hanson GN (1992) Boron isotopic composition and concentration in modern marine carbonates. Geochimica et Cosmochimica Acta 56: 537-543
Henehan MJ and 5 coauthors (2016) A new boron isotope-pH calibration for Orbulina universa, with implications for understanding and accounting for ‘vital effects’. Earth and Planetary Science Letters 454: 282-292
Howes EL and 8 coauthors (2017) Decoupled carbonate chemistry controls on the incorporation of boron into Orbulina universa. Biogeosciences 14: 415-430
Joachimski MM, Simon L, van Geldern R, Lécuyer C (2005) Boron isotope geochemistry of Paleozoic brachiopod calcite: implications for a secular change in the boron isotope geochemistry of seawater over the Phanerozoic. Geochimica et Cosmochimica Acta 69: 4035-4044
Kaczmarek K, Nehrke G, Misra S, Bijma J, Elderfield H (2016) Investigating the effects of growth rate and temperature on the B/Ca ratio and δ11B during inorganic calcite formation. Chemical Geology 421: 81-92
Klochko K, Cody GD, Tossell JA, Dera P, Kaufman AJ (2009) Re-evaluating boron speciation in biogenic calcite and aragonite using 11B MAS NMR. Geochimica et Cosmochimica Acta 73: 1890-1900
Klochko K, Kaufman AJ, Yao W, Byrne RH, Tossell JA (2006) Experimental measurement of boron isotope fractionation in seawater. Earth and Planetary Science Letters 248: 276-285
Legett S, Rasbury ET, Grossman EL, Hemming NG (2016) The boron identity: using δ11B to examine Late Paleozoic seawater. 2016 Goldschmidt Conference Abstracts. Geochemical Society, Yokohama, Japan, p 1720
Lemarchand D, Gaillardet J, Lewin É, Allègre CJ (2002) Boron isotope systematics in large rivers: implications for the marine boron budget and paleo-pH reconstruction over the Cenozoic. Chemical Geology 190: 123-140
Mavromatis V, Montouillout V, Noireaux J, Gaillardet J, Schott J (2015) Characterization of boron incorporation and speciation in calcite and aragonite from co-precipitation experiments under controlled pH, temperature and precipitation rate. Geochimica et Cosmochimica Acta 150: 299-313
Noireaux J and 7 coauthors (2015) Crystallographic control on the boron isotope paleo-pH proxy. Earth and Planetary Science Letters 430: 398-407
Penman DE, Hönisch B, Rasbury ET, Hemming NG, Spero HJ (2013) Boron, carbon, and oxygen isotopic composition of brachiopod shells: intra-shell variability, controls, and potential as a paleo-pH recorder. Chemical Geology 340: 32-39
Penman DE, Hönisch B, Zeebe RE, Thomas E, Zachos JC (2014) Rapid and sustained surface ocean acidification during the Paleocene-Eocene Thermal Maximum. Paleoceanography 29: 357-369
Ridgwell A (2005) A mid Mesozoic revolution in the regulation of ocean chemistry. Marine Geology 217: 339-357
Rollion-Bard C and 5 coauthors (2011) Boron isotopes as pH proxy: a new look at boron speciation in deep-sea corals using 11B MAS NMR and EELS. Geochimica et Cosmochimica Acta 75: 1003-1012
Sen S, Stebbins JF, Hemming NG, Ghosh B (1994) Coordination environments of B impurities in calcite and aragonite polymorphs: a 11B MAS NMR-study. American Mineralogist 79: 819-825
Simon L, Lécuyer C, Maréchal C, Coltice N (2006) Modelling the geochemical cycle of boron: implications for the long-term δ11B evolution of seawater and oceanic crust. Chemical Geology 225: 61-76
Stewart JA, Gutjahr M, Pearce F, Swart PK, Foster GL (2015) Boron during meteoric diagenesis and its potential implications for Marinoan snowball Earth δ11B-pH excursions. Geology 43: 627-630
Uchikawa J, Penman DE, Zachos JC, Zeebe RE (2015) Experimental evidence for kinetic effects on B/Ca in synthetic calcite: implications for potential B(OH)4− and B(OH)3 incorporation. Geochimica et Cosmochimica Acta 150: 171-191
Xiao Y and 5 coauthors (2006) An unusual isotopic fractionation of boron in synthetic calcium carbonate precipitated from seawater and saline water. Science in China Series B-Chemistry 49: 454-465