doi:10.1016/j.epsl.2006.05.001 
Copyright © 2006 Elsevier B.V. All rights reserved.
Chains, clumps, and strings: Magnetofossil taphonomy with ferromagnetic resonance spectroscopy
Robert E. Koppa, 
, 
, Benjamin P. Weissb, , Adam C. Maloofb, 1, , Hojotollah Valic, d, , Cody Z. Nasha, and Joseph L. Kirschvinka,
aDivision of Geological and Planetary Sciences, California Institute of Technology, Pasadena, CA 91125, USA
bDepartment of Earth, Atmospheric, and Planetary Sciences, Massachusetts Institute of Technology, Cambridge, MA 02139, USA
cDepartment of Anatomy and Cell Biology and Facility for Electron Microscopy Research, McGill University, Montréal, QC, Canada H3A 2B2
dDepartment of Earth and Planetary Sciences, McGill University, Montréal, QC, Canada H3A 2A7
Received 15 February 2006;
revised 26 April 2006;
accepted 1 May 2006.
Editor: S. King.
Available online 12 June 2006.
References and further reading may be available for this article. To view references and further reading you must
purchase this article.
Abstract
Magnetotactic bacteria produce intracellular crystals of magnetite or greigite, the properties of which have been shaped by evolution to maximize the magnetic moment per atom of iron. Intracellular bacterial magnetite therefore possesses traits amenable to detection by physical techniques: typically, narrow size and shape distributions, single-domain size and arrangement in linear chains, and often crystal elongation. Past strategies for searching for bacterial magnetofossils using physical techniques have focused on identifying samples containing significant amounts of single domain magnetite or with narrow coercivity distributions. Searching for additional of traits would, however, increase the likelihood that candidate magnetofossils are truly of biological origin. Ferromagnetic resonance spectroscopy (FMR) is in theory capable of detecting the distinctive magnetic anisotropy produced by chain arrangement and crystal elongation. Here we present analyses of intact and lysed magnetotactic bacteria, dilutions of synthetic magnetite, and sedimentary samples of modern carbonates from the Great Bahama Bank, Oligocene–Miocene deep-sea muds from the South Atlantic, and Pleistocene lacustrine deposits from Mono Basin, California. We demonstrate that FMR can distinguish between intact bacterial magnetite chains, collapsed chains, and linear strings of magnetite formed by physical processes. We also show that sediments in which the magnetization is likely carried by bacterial magnetite have FMR spectra resembling those of intact or altered bacterial magnetite chains.
Keywords: magnetotactic bacteria; biogenic magnetite; ferromagnetic resonance; magnetofossils
Fig. 1. Synthetic FMR spectra. Generated with g = 2.12, and (A) no magnetic anisotropy, (B) cubic Ban = − 55 mT, as expected for non-interacting cubic magnetite, (C) uniaxial Ban = 100 mT, (D) uniaxial Ban = − 100 mT, and (E) uniaxial Ban = 100 mT. For (A–D), σ = 25 mT; for (E), σ = 50 mT. Thick lines show derivative spectra and thin lines show integrated absorption spectra with Gaussian broadening lowered to σ = 3 mT. Sharp spectra with positive uniaxial anisotropy, as in (C), have two local maxima on the low field side, while sharp spectra with negative uniaxial anisotropy, as in (D), have two local minima on the high field side; these features can be obscured by spectral broadening, as seen in comparison of (C) and (E).
Fig. 2. Definitions of basic FMR parameters. Illustrated on a synthetic FMR spectrum of non-interacting, equidimensional magnetite (g = 2.12, σ = 30 mT, cubic Ban = − 55 mT, K2/K1 = 0.21). The dark line shows the derivative spectrum and the light line shows the integrated spectrum.
Fig. 3. Measurements of intact and altered AMB-1. (A) ARM acquisition curves, (B) ferromagnetic resonance spectra, and (C) low-temperature cycling curves of cultures of AMB-1. In (A), the lower dashed line is a chiton tooth standard for highly interacting magnetite. In (B), thick lines represent room-temperature measurements, thin lines represent 77 K measurements (where performed), and dashed lines indicate spectral fits.
Fig. 4. Measurements of synthetic magnetite. (A) ARM acquisition curves and (B) ferromagnetic resonance spectra of dilutions of synthetic magnetite powder TMB-100. In (A), the upper dashed line is intact AMB-1. The chiton tooth standard shown in Fig. 2a closely follows the line for sample T1a. In (B), thick lines represent room-temperature measurements, thin lines represent 77 K measurements, and dashed lines indicate spectral fits.
Fig. 5. Semi-log plots of the FMR parameters of the synthetic magnetite and AMB-1 samples against kARM/IRM.
Fig. 6. Transmission electron micrographs of Pt-C replica of synthetic magnetite powder TMB-100 diluted at 6000 ppm in sucrose. The bright material is the Pt-C replica of the sucrose dilutant, while the dark crystals are magnetite. (A) shows the association between clumps and strings of magnetite particles and the surface of sucrose crystals. (B), (C), and (D) show higher resolution images of the clumps and strings.
Fig. 7. Summary of FMR parameters. Plots of (A) effective g-factor vs. asymmetry ratio A, in the manner of Weiss et al. [20], and (B) full width at half maximum (ΔBFWHM) vs. A. In (A), the dashed lines demarcate the area defined by Weiss et al. [20] as being the domain of magnetosome chains, with A < 1 and geff < 2.12. In (B), the dashed lines are contours of constant α values, where α is defined in the text. Red and green arrows highlight the dilution trends for, respectively, SDS-treated AMB-1 and TMB-100. Parameters for Geobacter metallireducens GS-15, which precipitates extracellular superparamagnetic magnetite, and for the magnetotactic bacteria strains MV-1, MS-1, and MC-1 (arranged in order of increasing A) are derived from the data of [20].
Fig. 8. Measurements through Andros Island core C51. (A) Smoothed ferromagnetic resonance spectra, (B) ARM acquisition curves, and (C) plots of FMR parameters against kARM/IRM. In (A), the small peak at 
160 mT is the paramagnetic resonance of Fe+3 and the jaggedness at 330 mT is the remanent of the Mn+2 resonance, most of which was removed by smoothing. On (B), the upper dashed line is the ARM acquisition curve of intact AMB-1, while the lower dashed line is that of the highly interacting chiton tooth standard. The 31 cm ARM curve, not shown, nearly follows the 14 cm curve.
Fig. 9. Representative FMR spectra of (A) Oligocene–Miocene deep-sea muds from DSDP Leg 73 Site 522, in which magnetofossils are a major remanence carrier and (B) silts from the Pleistocene Wilson Creek Formation of Mono Basin, which are magnetically dominated by detrital magnetite. In (A), the dashed line shows by way of comparison AMB-1 sample A3c, which was treated with SDS and diluted in sucrose by powdering for 5 min.
Fig. 10. The factor α for model spectra. α is plotted for synthetic spectra with first-order uniaxial anisotropy specified by Ban. Contour lines show values of σ ranging from 20 mT to 120 mT in 10 mT steps. In the region occupied by measured magneotactic bacteria and magnetofossil-bearing samples (Ban ≥ 75 mT, α < 0.30), α varies nearly linearly with σ.
Table 1.
Summary of FMR and magnetic parameters of bacterial samples
Table 2.
Summary of 77 K FMR parameters of bacterial and synthetic magnetite samples
Table 3.
Summary of FMR and magnetic parameters of synthetic magnetite samples
Table 4.
Summary of parameters of environmental samples
Abstract
Purchase PDF (793 K)
E-mail Article
Add to my Quick Links
Cited By in Scopus (7)
Purchase PDF (35 K)
