Roman Pozzolana and mortar dating C: Applications and experiments 29 samples CI 2 inconclusive samples CI-CII 19 conclusive samples CIV 7 samples fulfill less secure Criteria III and IV CIII 1 inconclusive sample Second fractions agree in 12 samples from Trajan’s Market and Ostia B: No age control A: Age control 0 100 200 300 400 500 600 1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 All mor- tars Lime mor- tars Pozzo- lana Lime lumps H3PO4 HCl Clas- sical Diagram of all mortar samples analyzed Type Hydrolysis Chronology Medi- eval This poster focuses on recent methodological development in dating mortar, exspecially on classical archaeology and the dating of lime lumps embedded in the mortar Mortar is not an organic material. Yet the chemical process in the hardening in principle makes it into an ideal matrix for 14 C dating. To make mortar, limestone has to be heated up to at least 900 ºC. After the carbon dioxide has been released in the process, calcium oxide (un-slaked lime) is mixed with water and aggregate, usually sand. In the hardening process the slaked lime reacts with atmospheric carbon dioxide, and calcium carbonate is produced. Thus the mortar absorbs the carbon dioxide from the atmosphere and thereafter behaves as if it were organic. Unburned limestone in the mixture can contaminate the result, yielding dates too old. Diagram displaying mortar samples analyzed. By now the databank covers results from more than 550 analyses of some 470 individual samples. Some of the samples were analyzed repeatedly, in different laboratories. The samples include different sorts of mortar, and they cover both Classical and Medieval archaeology. Recent Advances in Radiocarbon Dating of Ancient Roman Mortar and Concrete Lime lumps embedded in mortar In the process mortar is more easily soluble than limestone, thussthe first fractions should be free from contamination. Therefore, to illustrate the entire dissolution process, the gas-flows interrupted at several stages create a sequence of CO 2 age profiles. The contamination is shown towards the end of the profile. 2 1 3 4 5 5 CO 2 -fractions 0 100 CO 2 in the sample (%) 14 C-age (BP) To avoid the effects of contamination different types of separation are needed. At first a mechanical separation with gentle crushing of the sample. The grains are then sieved into 39/45-75 microns for further analysis. In the following chemical separation 85% of phosphoric acid is poured over the sample, creating a gas of carbon dioxide. LIQUID NITROGEN CO2 FRACTIONS COLD TRAP SAMPLE PRESSURE GAUGE VACUUM PUMP Åsa Ringbom, Åbo Akademi University, Finland John R. Hale, University of Louisville, Kentucky Torre delle Milizie, Rome, The result of Rome 007Li, (Li=lime lump) as compared to analysis of Rome 007 from different AMS laboratories (Aarhus, Denmark, Oxford UK, Tucson Arizona). The lime lump analysis supports our earlier interpretations - that the top of Torre delle Milizie was erected towards the end of the 13 th century. Santa Costanza, at Via Nomentana, Rome. The age of construction is debated. It was built as a mausoleum either by Constantina, daughter of Constantine the Great, before 350, or by her brother–in–law Julian the Apostate some 10-12 years later. Mortar samples were analyzed in Oxford, Tucson and Aarhus all forming confusing age profiles. The lime lump Rome 042 was analyzed twice, resulting in an almost horizontal age-profile, which is easy to interpret. But because the calibration curve wiggles at this point the result is not helpful enough in pinpointing the chronology. We can never defeat the irregularities of the calibration curve. Conclusions: a) Medieval lime mortars in Scaandinavia and Classical mortars based on marble (Torre de Palma) have yielded promising results. b) Hydraulic Pozzolana mortars have been less successful. c) Pozzolana mortars from Pompeii and Herculaneum cannot be dated, regardless of whether bulk mortars or lime lumps embedded in the mortars are used. d) Lime lumps embedded in the mortar have shown a great deal of success, both in lime - and pozzolana mortars. Lime lumps generally contain very little contamination from unburnt limestone, thus fewer CO 2 fractions are needed, thus cheaper. e) Above all, for pozzolana mortars analyzing lime lumps may be the obvious solution. Why mortar? The aim of archaeological excavations is often to establish the chronology of a site. Mortar, lime mortar or pozzolana mortar, is often the only datable material available, in abundance, from all the different building stages. Differently from other datable materials like dendrochronology and 14 C of organic materials, and differently from archaeological artifacts such as coins and ceramics, mortar analysis dates the actual time of the construction, or when the mortar hardened. It would therefore be helpful to use mortar dating on a routine basis in archaeological excavations or in a process of conservation. The Roman Villa of Torre de Palma, Portugal, mortar analyzed in 1997-2000, was our first encounter with Classical archaeology. Out of a total of 64 samples analyzed 18 yielded Criterion I results. Independently of other materials and methods these 18 samples could indicate the date of 12 out of 21 structures, thus marking the outlines of the chronology of the site. The earliest structure was the temple in the East Court (70-170AD), and the latest are the apse 3 of the Basilica (580-630AD) and the large font of the Baptistery (570-690AD). Reliability criteria for the interpretation of the age profiles Criterion I: when analyzed in two to five CO 2 fractions and the results from the two first CO 2 fractions agree Criterion II: when there is mutual agreement between the dates of the first CO 2 fractions in a seriesof three or more samples from one single building. Criterion I and II yield a conclusive date independently of age control. Dating medieval lime mortars in Scandinavia has yielded promising results.More than 96% if all the results agree with age control from other dating methods, as shown from the Åland Islands between Finland and Sweden Dating pozzolana mortars, however, was problematic. Only some 50% of all samples could be considered successful, and the interpretation of the CO 2 age profiles was seen as complicated. A Mortar Dating Workshop in Åland, 2006: Fiona Brock, John R. Hale, Jan Heinemeier, Åsa Ringbom, Alf Lindroos and Pia Sonck-Koota. The International Mortar Dating Project 1994 - Map of mortar dating sites, those in red mark lime mortars, those in green pozzolana mortars. 200 400 0 km N CII Conclusive results 45 Inconclusive results 26 CI Not agreeing 4 Mutually agreeing results 75 CII CIII CIV Age control, 79 samples No Age control, 71 samples Dendro Och 14CW 14C W Dendro and 14C W Dendro CI CI-CII CII CI-CII CI CIII, Too few samples per building- unit. Age- Profiles cannot be interpreted The mortar dating group, Lynne Lancaster, Jan Heinemeier, Åsa Ringbom and Alf Lindroos outside the Colosseum, where successful results meet the demands of Criterion I and Criterion and II, and the calibrated result agrees with written sources. 1800 1900 2000 2100 2200 2300 2400 2500 2600 2700 2800 2900 0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 F C-14 age BP Colosseum 003, 76-150 m 001, 46-75 m 002, 46-75 m Heavily contaminated No re-crystallization First fractions converge and are correct Atmospheric data from Reimer et al (2004);OxCal v3.10 Bronk Ramsey (2005); cub r:5 sd:12 prob usp[chron] 200CalBC 100CalBC CalBC/CalAD 100CalAD 200CalAD 300CalAD Calibrated date 1800BP 1900BP 2000BP 2100BP n o i t a n i m r e t e d n o b r a c o i d a R The Colosseum : 1940±17BP 68.2% probability 25AD (13.4%) 40AD 45AD (54.8%) 80AD 95.4% probability 10AD (86.5%) 90AD 100AD ( 8.9%) 130AD Why lime lumps? Lime lumps embedded in the mortar are highly interesting in this context. There may still be clods of quicklime in the putty, which become slaked and carbonated possibly already before the aggregate is added. These form aggregate free white lumps in the mortar. In principle therefore well-burnt lime lumps should be free from contamination. CO 2 CaCO 3 limestone CaO unslaked lime Ca(OH) 2 slaked lime + H 2 O water carbon- dioxide Statistics of results from 32 liumps embedded in the mortar. 25 lime lumps were identified in lime mortars, whereas 7 were embedded in pozzolana mortars. Of these 32 samples, 7 had been exposed to fire. The majority comes from medieval Scandinavia. 0 5 10 15 20 25 30 35 40 1 2 3 4 5 6 7 1 2 3 4 5 6 7 All lime lumps success rate Nordic medieval Classical Pozzo- lana Rome Pozzolana Pompeii Torre de Palma Current statistics of lime lumps Terracina, Temple of Anxur. Repeated analysis of a bulk sample of pozzolana mortar (Terracina 001)resulted in two different looking age profiles, which seemed to converge at the first CO 2 fractions. A lime lump embedded in the pozzolana confirmed the earlier interpretation. 500 700 900 1100 1300 1500 1700 1900 2100 2300 2500 0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 C-14 age BP F Rome, Torre delle Milizie Oxford 76-150 m H 3 PO 4 Sample Rome 007 Tucson 46-75 m H 3 PO 4 Tucson 46-75 m HCl Århus Lime lump 21-150μm H3PO4 Århus 46-75μm H3PO4 CO 2 CaCO 3 CaO Ca(OH) 2 +sand + H 2 O + H 2 O 14 CO 2 from atmosphere Samples from Pompeii or Herculaneum Delayed hardening Experiments Crushed bricks sec. opus floors Too few fractions analyzed Re-crystallization with old layers of stuccho 200 45-80 10-22 6-3 Atmospheric data from Reimer et al (2004);OxCal v3.10 Bronk Ramsey (2005); cub r:5 sd:12 prob usp[chron] Torre de Palma - Dates based on agreement between 1st and 2nd fractions in CO2 600BC 400BC 200BC BC/AD 200AD 400AD 600AD 800AD Calibrated date The Temple in East Court AAR-3933 TP 108 Cat1 1885±35BP South Field West AAR-5441 TP 186 1683±45BP AAR-5442 TP 188 1750±45BP south west house AAR-3929 TP 069 1711±35BP The Baptistery, earlier structure AAR-10361 TP 013 1684±35BP AAR-3411 TP 201 1738±55BP Precinct wall AAR-5440 TP 182 1601±35BP The Olive Press AAR-4231-2 TP 083-2 1628±55BP The Basilica, phase 2 AAR-3926 TP 023 1535±30BP AAR-4825 TP 143 1554±20BP Basilica, Apse 4 AAR-5653 TP 209 1483±30BP North East House AAR-5428 TP 045 1543±35BP AAR-5643-1 TP 178-2 1515±45BP Maosoleum AAR-4228 TP 036-2 1576±45BP Basilica, Apse 3 AAR-3927 TP 025 1410±35BP AAR-3928 TP 026 1448±30BP AAR-3414 TP 206 1417±30BP Babtistery, large font, second stage AAR-4822-3 TP 138-3 1393±40BP North Barn SOUTH FIELD WEST Southwest Cemetery (CWR) Northwest Cemetery BASILICA NORTHEAST BUILDING Garden Q uarter Peristyle House East Bath Forge Atrium House East Court South Hall West Court Gate Gate Gate Olive Press Portico House West Bath Southwest House A Southwest House B (CO MPLEX B) (CO MPLEX C) (CO MPLEX R) South Field TORRE DE PALMA Location of Studies Samples 23 25 26 1 013 138 143 4 182 10 m 209 144,2 19 145 211 147 175 32 199 139 TORRE DE PALMA Basilica (Complex B) Criterion I (black) Criterion II (red) 17 144,1 1800 2000 2200 2400 2600 2800 3000 0 0,2 0,4 0,6 0,8 1 Terracina, temple of Jupiter Anxur 001; 46-75μm 001Li; 21-150μm F Sample 001 Lime lump H 3 PO 4 , Århus Bulk mortars 14 C age BP H 3 PO 4 (Århus) HCl (Tucson, Arizona) 1300 1400 1500 1600 1700 1800 1900 2000 2100 2200 2300 2400 0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 St Costanza, Rome C-14 age BP F Samples Rome 042 Rome 017 Rome 046 46-75μm fracons Rome 042Li 21-75 and 76-150μm (1 fracon; 3 fracons) 046 017 042 Lumps: 1716 15 320-390 (42,6% 1σ) +/- Terracina 001Li R_Date(1909,35) 95.4% probability 19 (95.4%) 214calAD 200 100 1calBC/1calAD 101 201 301 401 Calibrated date (calBC/calAD) 1600 1800 2000 2200 Radiocarbon determination (BP) OxCal v4.1.7 Bronk Ramsey (2010); r:5; Atmospheric data from Reimer et al (2009);