Part I. FTIR Results - Francesco Berna

Introduction

       This work focuses on the silicate transformation occurred in the combustion features. In fact, sediments exposed to high temperature are structurally modified. This modification occurs if an open fire is made directly on the sediments or if sediment adhering to the fuel is incorporated into the fire. Identifying the temperature range at which a sediment particle was burned is largely based on characterizing the irreversible transformations that occur due to the firing of silicate minerals, and in particular clay minerals. Clay minerals undergo several not-reversible temperature-specific structural and compositional changes when exposed to increasing temperatures. In a recent study after performing heat experiments with standard materials and control sediments we showed that silicate alterations due to high temperature can be identified (mainly by FTIR spectroscopy) in the sediments of archaeological sites and used for the study of human activities and site formation processes (Berna et al., 2007). In this NSF-funded research I am applying a similar approach and trying to determine the temperature range of heated sediments composing the combustion feature excavated at Kebara, Hayonim and Pech de l’Azé caves. In addition I am developing a protocol to identify by FTIR spectroscopy heated silicates (and their temperature) directly in thin sections of combustion features. Particles embedded in uncovered thin sections can be directly analyzed by FTIR either using an ATR device or an IR microscope equipped with an ATR objective (Figure 1). Both devices use ATR crystals that by touching the particles will allow the acquisition of IR spectra. Particles with a diameter as small as 100 m can be analyzed with an ATR objective and larger particles (5mm diameter) with an ATR device. Unfortunately spectra of mixed phases, such as sediments are not directly searchable with common IR spectra reference databases. The latter IR spectra in fact are collected by suspending the material in KBr pellets. Moreover the absorptions of the embedding material (epoxy or polyester) interfere with the absorptions of the sedimentary material. It is therefore important to create an ATR database of burned sediments and standard materials embedded in thin sections. My work is subdivided in several tasks as follows:

Mineralogical Characterization of Control Soils from Kebara Cave Israel

Methods
• Clay size fraction separation by pipette method
• FTIR spectroscopy
• XRD

Results
• Control sediments (not altered by fire or diagenetic processes) were collected several meters above the cave chimney and in the cave. The clay component of these control sediments contains smectite (probably montmorillonite), illite and significant amounts of kaolinite (Figure 2).

In Progress
• Further investigation on the clay mineralogy by XRD.

Heating Experiments with Kebara Control Sediment and Standard Clay Minerals

Methods
• Heating experiments: Aggregates of control sediments collected at Kebara and of standard clay minerals (montmorillonite, illite, and kaolinite) were heated for 2 hours in a muffle furnace at temperature between 300 and 1100 °C (Figure 3).
• Bulk analyses: The heated aggregates were sub sampled for powder analyses by:
  • FTIR
    • Suspended in KBr pellet
    • ATR attachment
  • Thin Section preparation
    • The heated aggregates are embedded in polyester resin. Thin sections of this set of samples will be made shortly.

Results
• The rubefication of the sediment is evident at 400 °C. Characteristic changes in the IR patterns of the heated sediment start to show at 350 °C, especially in the OH stretching area (~3700-2800 cm-1). As expected, at and above 450 °C changes in the IR spectral patterns become very evident in the major Si-O-Si absorption that gradually shifts towards higher wavenumbers with increasing heat. Although with slightly different absorption values, these gradual spectral changes are detected both in samples analyzed by suspension KBr pellets or in contact with ATR device. We therefore have FTIR databases (KBr pellet and ATR) of Kebara heated sediments (Figure 4). We can use these reference data bases to determine if any of the sediments composing the combustion features are burned and to determine their burning temperature.

In Progress
• Preparation of the thin sections of heated aggregates and standard materials.
• ATR analyses on thin sections of heated aggregates and standard materials
• XRD and Raman of heated aggregates and standard materials.

Analyses of the Combustion Features Sampled at Kebara Cave Israel during the 2006 Excavation Campaign

Sampling in the field:
• 28 combustion features were identified and described in Kebara cave. Blocks of the structures were collected for micromorphological analyses and loose samples of the different constituent sub-layers identifiable in the fields were sampled for bulk analysis.

Methods
• FTIR analyses of the bulk and single components of the sub-layers composing the structures
• Thin sections

Results
• The IR analyses performed so far show that the diagenesis of the combustion structures sampled in 2006 is of modest intensity. In fact, the calcite of the ashes is either preserved or slightly phosphatised into carbonated apatite. The analyses also show that the silicates of some of the combustion features of the Mousterian layers were burnt at 450-500°C, with no sign of higher temperatures. Interestingly the silicates of one of the Upper Paleolithic combustion features (KEB06-03) have IR absorptions at 890 and 1070 cm-1 and are not easy to interpret. Absorptions at 1070 cm-1 are characteristic of high temperature glasses and/or obsidian. It is also possible that these IR absorptions are caused by significant concentrations of phytoliths and diatoms normally contained in these sediments. This aspect needs to be investigated.

In progress
• Loose sediments: complete the FTIR analyses
• Thin sections:
  • Micromorphological analyses
  • FTIR microspectroscopy

Definition of a Protocol for the Infrared spectroscopy Analysis of Thin Sections of Combustion Features

Methods
• I performed preliminary analyses with the ATR device and the ATR microscope objective of several particles embedded in thin sections of combustion features. In slides KEB96-12b and HAY 98 202 (figure 5), I managed to analyze the same particles with the two methods and compare the spectral patterns.

Results
• The IR spectra obtained with the ATR device of discrete particles embedded in thin sections are well reproducible. The peaks are less sharp as compared to the spectra of similar loose material, due to the interference of the absorption of the embedding resin. The spectra obtained with the ATR objective show noisier spectra and lower reproducibility as compared to the ATR device. The reduced quality is probably due to the smaller area of contact and lower effective IR radiation exciting the samples in the ATR objective. Furthermore the Si-O-Si absorption acquired with ATR objective shows slight shifts with respect to the spectra acquired with ATR device (Figure 6). Nevertheless the comparison of these spectra with the IR reference database of burned sediments suggests that the sediment particles analyzed contain silicates burned at temperature above 400°C.

In progress
• Acquisition of reference database of thin sections of experimentally burned sediments and standard materials
• Assess the effect of thin section polishing to increase the quality of the spectra.

Figures

Figure 1: FTIR system. A) General view of the FTIR bench (Nexus 470) and the IR microscope (Continum); B) Detail of the ATR objective and thin section under investigation; C) Detail of the ATR crystal of the ATR device; D) Thin section being analysed with the ATR device

Figure 2: FTIR spectra (KBr pellet) of soil collected in the proximity of the chimney of Kebara cave. Legend: K=Kaolinite, I/S= Illite Smectite, Qz= Quartz

Figure 3: Burning experiments. A) Samples in muffle oven, B) Unheated and heated control sediment from Kebara cave

Figure 4. FTIR spectra of unheated and heated control sediment collected with ATR device.

Figure 5: Thin sections of combustion features

Figure 6: FTIR spectra of orange-brown particle (5mm) in thin section (HAY98-202_3). Spectra are collected with ATR objective (bottom red) and ATR device (top blue).

References

• Berna F., Behar A., Shahack-Gross R. , Berg J., Boaretto E., Gilboa A., Sharon I., Shalev S., Shilstein S, Yahalom-Mack N., Zorn J. R., Weiner S. (2007) “Sediments exposed to high temperatures: reconstructing pyrotechnological processes in Late Bronze and Iron Age Strata at Tel Dor (Israel)” Journal of Archaeological Science 34, 358-373