Carlsbergfondet - Hand-held X-ray fluorescence (HHXRF) for trace metal analysis

Projekter: ProjektForskning



All elements of the periodic table can be found dissolved in seawater, with concentrations varying from element to element. These elements are removed by a variety of processes, some of which are specific to the chemical environment. For example, the chemical forms of some of the dissolved metals like uranium, molybdenum, vanadium and rhenium are sensitive to the presence or absence of oxygen, and these metals are preferentially removed from seawater and concentrated in sediments depositing in oxygen-free (anoxic) environments (Jacobs and Emerson, 1982; Morford et al., 2009). Indeed, for molybdenum and many other transition metals like copper, nickel, zinc, and lead, the removal is enhanced in the presence of sulfide (Jacobs and Emerson, 1982). Therefore, the concentrations of trace metals as preserved in modern sediments (Boning et al., 2009) and in ancient sedimentary rocks (Algeo, 2004) provide strong indications of the chemical environment where the sediments deposited. In this way, ancient oxygen-poor environments like oxygen-minimum zones and sulfidic water bodies (the Black Sea is a modern example) can, in principle, be identified (Algeo, 2004; Boning et al., 2009).

Such oxygen-free marine environments were likely more plentiful in the geologic past when atmospheric concentrations were lower (Canfield, 1998; Raiswell and Canfield, 2012). Just how plentiful, however, is still unclear. It is clear, however, that an expansion of anoxic environments should reduce the concentrations of many trace metals in the oceans due to more active removal. This is important because many of the trace metals in seawater are critical components of enzymes conducting key biological processes like nitrogen fixation, denitrification and methanogenesis. Therefore, if trace metal availability becomes limited due to accelerated removal, then the biological processes requiring these trace metals may also become limited, influencing possibly, global biogeochemical cycles and even the evolution of life (Anbar and Knoll, 2002; Konhauser, 2009). Thus, an accurate accounting of trace metal concentrations in sediments can inform us as to both the evolution of ocean chemistry through time, and the influence of ocean chemistry on trace metal availability.

Traditionally, trace metal analyses have been conducted with either inductively coupled plasma in line with a mass spectrometer (ICP-MS), or by standard X-ray fluorescence (XRF). Both of these techniques require a great deal of sample preparation and are extremely expensive, providing a serious limitation on the number of samples that can be analyzed. Recently, developments in hand-held X-ray fluorescence units (HHXRF) have reached sufficient sensitivity and accuracy to provide a reasonable, and inexpensive, alternative to the traditional methods of trace metal analyses (Dahl et al., 2013). The initial cost of the machine is the largest expense, with virtually no running costs. In addition, sample preparation requires only powdering of the samples. In addition, each analysis takes only a few minutes, so that sample throughput is high.

Funds are requested to purchase such an instrument. The instrument will be used to complement ongoing work in the lab focused on deciphering marine chemical environments in the geologic
past, and on determining the history of trace metal concentrations in the oceans. This instrument will also provide a platform for seeking funds to further this work. This is relevant, as the lab has for many years been supported by substantial funding from the Danish National Research Foundation, but the support ends in July, 2015. The ERC also supports projects related to the oxygen-regulation of microbial metabolisms, but trace metal analyses is outside of the focus of this project.


With the aim to measure the elemental composition of scientific rock samples, in a non-destructive, simple and environmentally sustainable way, we have first calibrated a protocol to ensure high-accuracy interpretation of the data, and then used in on unknown samples with satisfactory results.

To first develop a protocol to ensure that we interpret the data with high accuracy and precision, we measured a dozen certified rock standards and twice as many real samples (Z4, U4 and CJ) with both the HHXRF and – for true values – with ICP-MS and XRF. Our ambition was to create calibration curves for each element. A few challenges needed to be addressed. For example, we had to explore discrepancies in the data, when measured with either of the two software modes Soil and Geochem, that adjoin the HHXRF gun. P showed significantly higher precision when measured in the Geochem mode than in the Soil mode (Figure 1), while Ni was about equal (Figure 2). We now trust the Geochem mode to generally measure major elements with high precision, such as our key element Fe (Figure 3), and the Soil mode to, generally, measure trace elements with high precision.

We also identified a cryptic drift in the data over time, which could not be explained with any physical properties of the sample (e.g. moist uptake) or container (e.g. increasing dust on static plastic film), but which was eventually solved by a software update at the manufacturer after repeated communication with them. Now, after additional testing, it is safe to say that the instrument function well and that we are able to interpret the data with the highest precision and accuracy.

Taken together, the calibration process alone has learned us how to take care during measuring (materials used), to optimize measuring conditions for highest precision (repeats and software mode), to see the limits of detection level, and how to calibrate the data.

Our use of the HHXRF has already assisted us in understanding Earth settings in deep time. We have, for example, identified the elemental composition of rock samples from the 1.4-billion-yearold Xiamaling formation. The Xiamaling is a peculiar site and formation, with for example clearly rhythmical deposition of two significantly different sediment types (Figure 4).

Our group has, in collaboration with Chinese team, described that the rhythmical alterations stem from climate change that was paced by orbital forces in our Solar system. Evidence of climate change this long ago, is the earliest yet described, where the chemical evidence of alternations plays a key role. Also, it has been a privilege to describe that interactions between the tilt, orbit and spin of Earth, Jupiter and the moon indeed affect water column ecology and conditions at particular site more than a billion years ago.

The chemical evidence at Xiamaling – in part retrieved with the HHXRF – has also revealed that the ocean at that time contained more dissolved oxygen than animals require. This interpretation is built on for example vanadium, molybdenum and iron, as well as carbon and hydrogen data, and adds to the debate on whether an increase of atmospheric oxygen triggered the diversification of animals some 800 years later.

We continue to explore the Xiamaling Formation, using the HHXRF, where particularly an iron rich unit is peculiar. We will also explore more recent rock samples, such as from the Silurian and Devonian when plants colonized land and, likely, changed Earth’s surface environment significantly among other scientifically relevant sites.
Effektiv start/slut dato01/01/201529/02/2016