Large-scale 2nd to 3rd century AD bloomery iron smelting in Korea

https://doi.org/10.1016/j.jas.2010.12.007Get rights and content

Abstract

Iron production in Korea has traditionally been seen in the shadow of developments in cast iron technology in China, with limited indication for a northern influence via Russia’s Maritime Province. The possibility of the existence of bloomery iron production in ancient Korea has been little explored, and relevant discussion is fraught with speculations based primarily on the early use of cast iron. The recent excavation of a site in South Korea recovered substantial amounts of slag providing direct evidence of bloomery smelting. The accelerator mass spectrometric dating of burnt wood from inside one of the slag pieces showed that the site was in use in the early 3rd century AD or earlier, which is in agreement with the assessment based on ceramic typology. The traits of a bloomery process evident in the slags’ microstructure, shape, composition and excavation context are discussed along with the implications for historical iron technology in Korea, where cast iron and the influence from China have been overly emphasised.

Highlights

► Slag from a Korean iron site was examined for microstructure and composition. ► Radiocarbon age by AMS placed the date in the early 3rd century AD or earlier. ► Evidence of an established bloomery smelting technology was found. ► Contact with the north is proposed as a probable source of early Korean iron. ► Questions are raised to the theory of Korean iron originating purely from China.

Introduction

Cast iron objects excavated from some of the earliest iron-yielding sites in Korea are frequently cited to claim that the Korean Iron Age began under a strong influence from China (Kim, 1998: 101–103). This premise was further supported by the fact that the early cast objects were in many cases recovered along with the knife-shaped Mingdao coins from the northern Chinese state of Yan that lasted from 323 to 222 BC. It is generally accepted, therefore, that the Korean Iron Age commenced at around 300 BC and was established on a technological framework similar to that of China. Chinese iron technology at the time was characterised by the smelting of cast iron and the subsequent treatments of cast iron for other iron and steel products. According to Wagner (1996: 289) and Rostoker and Bronson (1990: 149) evidence known of the practise of a bloomery process in ancient China is sparse and far from being conclusive (Wagner, 2008: 88–89, 104–105).

Taylor (1989) summarised the discussion concerning Korean early iron use based on the evidence available to her, which favoured a Chinese dominance in the introduction of iron. Unfortunately, little evidence for iron smelting in Korea was available at the time, and Taylor (1989) remains vague about the nature of the suggested local production: indirect cast iron or direct bloomery iron smelting. Among the objects, white cast iron and decarburised steel seem to dominate over steel made from carburised bloomery iron. This, of course, implies the existence of some bloomery smelting, and she cites work by Lee (1982) and Yun (1984) reporting small-scale smelting sites, often near the Han and Nakdong rivers in the south.

Contrary to the general consensus on the origin of Korean iron, Kim (1977) proposed another theory based on the work done in North Korea reporting the occurrence of iron artefacts at sites with strong Siberian traits. These sites are located at the northeastern part of the Korean peninsula close to the Maritime province of Russia, and are dated to a period several centuries earlier than those showing the earliest Chinese influence of the 4th to 3rd century BC. This led him to regard contact with the north as another probable route of iron introduction into Korea. The very early dates suggested by North Korean scholars are often considered to be motivated by nationalism. But it is of significance that two cases of the earliest iron reported were found at the handle of a bronze sword and at the tail of a bronze arrowhead (Kim, 1977). This indicates that the use of iron was just beginning, and the new metal was not chosen for strength. In addition, the sword was excavated from a stone cist tomb of Siberian style, a typical burial form of the Korean Bronze Age. A few artefacts do not signify the beginning of the Iron Age, but they do indicate the presence of iron in Korea before the influx of Chinese technology. Even though Kim (1977) was not aware of the technological implication of his postulate, this early iron from the north could not be the Chinese-style cast iron but it must have been derived from an older bloomery process.

An iron industry with bloomery smelting would be different in its origins and cultural implications from one without it, and this alternative should seriously be taken into account if the history of iron use in ancient Korea is to be properly understood. Unfortunately, most papers on the origin and development of Korean iron technology do not even distinguish between the two iron smelting processes, which are fundamentally different in their historical and technological contexts. Previous researches depended heavily on object typologies and were biased towards the theory of a Chinese origin. Furthermore, evidence thus far proposed for the existence of a fully developed bloomery process was not conclusive enough to claim another route of technological transmission. In this respect, the large amount of slag excavated from an ancient iron-making site at Yonjeri in the district of Cheongwon in Korea and the context from which they came prove to be rare and valuable evidence confirming the early practise of bloomery smelting in the Korean heartland. The present study will characterise their microstructure, chemistry and chronology by employing optical microscopy, scanning electron microscopy (SEM) equipped with energy dispersive spectrometry (EDS), X-ray diffractometry (XRD), X-ray fluorescence analysis (XRF) and accelerator mass spectrometry (AMS).

Section snippets

Comments on the site and artefacts

The asterisk in Fig. 1, a political map of the Korean peninsula around the 3rd century AD, locates the site from which the slag objects examined were excavated. The map shows the four contemporary rival kingdoms, Koguryo, Paekche, Silla and Kaya, along with their respective political centres at the time, Jiban, Seoul, Kyongju and Kimhae. The site is within the former Paekche territory.

The site of Yonjeri (Jo et al., 2008) is situated on a gentle slope of a low hill at the edge of the flood

Instrumental examination

The three groups of slag visually identified were further studied by microscopic and chemical analysis. This showed that the flow slag (Fig. 3a) and the blocky slag (Fig. 3c) were similar in their microstructure and chemistry, while the magnetic fragments shown in Fig. 3b form a separate group, interpreted as partly-reduced ore.

Accelerator mass spectrometric (AMS) dating

The samples for radiocarbon dating were taken from burnt wood buried inside a slag object and were sent for AMS measurement to the University of Arizona’s NSF-Arizona AMS Facility for 14C analysis in the USA and also to the Christian-Albrechts-University’s Leibniz Laboratory AMS Facility in Germany. The results are summarised in Table 2 where the calibrated or calendar ages were calculated using Calib Rev 5.0.1 (Stuiver and Reimer, 1993; Stuiver et al., 2005) and the data set IntCal04.14c (

Discussion

Choi and Yun, pioneers in Korean archaeometallurgy, acknowledge China as the major source of Korean iron and, at the same time, accept the existence of a bloomery smelting process in ancient Korea (Choi et al., 1998a, Choi et al., 1998b, Yun, 1984). They are in disagreement, however, on how low carbon iron was produced. Choi (2000) claims that bloom smelting was the major technique to produce low carbon iron, denying the practise of any Chinese-style finery in pre-industrial Korea, as opposed

Conclusion

Evidence of substantial bloomery smelting has been found in the appearance, microstructure, composition and excavation context of slag from the Korean iron-smelting site at Yonjeri. The site is located within the former territory of the Paekche kingdom near the modern city of Cheongju, approximately 100 km to the south of Seoul. Radiocarbon measurements by AMS place its date in the early 3rd century AD or earlier, consistent with the relative chronology for the site obtained based on pottery

Acknowledgement

The authors would like to thank the Jungang Research Centre of the Cultural Heritage for providing the slag objects examined in this work. The support from Dr. S.M. Shim and his students in the Ceramic Department of Hongik University is acknowledged for the XRD analysis and the fine polishing of ore specimens. This work was supported by the Korea National Research Foundation (NRF-2010-0028714). Detailed comments by five anonymous reviewers helped to considerably improve the manuscript and are

References (30)

  • R.B. Gordon

    Process deduced from ironmaking wastes and artefacts

    Journal of Archaeological Science

    (1997)
  • N.O. Kozhevnikov et al.

    An accidental geophysical discovery of an Iron Age archaeological site on the western shore of Lake Baikal

    Journal of Applied Geophysics

    (2001)
  • J. Choi et al.

    On the bloom and slag unearthed from Uksoo-dong Taegu-city of the 6th century

    Journal of the Society of Korean Traditional Science and Technology

    (1998)
  • J. Choi et al.

    On the nail unearthed from Bubheung-ri, Pajoo of the 15th century

    Journal of the Society of Korean Traditional Science and Technology

    (1998)
  • Choi, J., 2000. Was the Chao process ever practiced in Korea? In: Proceedings of the 11th conference, Korea Society for...
  • S.G. Jo et al.
    (2008)
  • D. Killick et al.

    Microstructures of puddling slags from Fontley, England and Roxbury, Connecticut, USA

    Historical Metallurgy

    (1987)
  • Kim, J.B., 1977. Iron culture in Korea. Study on Korean History 16, 3–27 (in...
  • J.G. Kim et al.

    The Great HwangNam Tomb: Excavation Report on the South Tomb (Text and Photos)

    (1994)
  • W.Y. Kim

    Introduction to Korean Archaeology

    (1998)
  • N.K. Lee

    A study of early iron age culture in South Korea

    Hankuk Gogo-Hakbo

    (1982)
  • E.M. Levin et al.

    Phase Diagrams for Ceramists

    (1964)
  • M. Ozawa

    Metallurgical Examination of the Objects Excavated from an Ironmaking Site at Milyang Sachon. Iron-making Site at Milyang Sachon, Excavation Report 1

    (2001)
  • J.S. Park

    On the Iron-making Wastes Excavated from a Site at Kwongokdong in Ahsan and the Process Involved. The Historical Site at Ahsan Kwongokdong, Excavation Report 77

    (2006)
  • Park, J.S., in review. Plate-type iron artefacts; their role and implication in the ancient iron technology of Korea....
  • Cited by (23)

    • The earliest box-shaped iron smelting furnaces in Asia: New data from Southern Siberia

      2020, Journal of Archaeological Science: Reports
      Citation Excerpt :

      And according to one more perspective, Russian Maritime Province was a source of knowledge for Korean iron production (Park and Rehren, 2011). However, archaeological data show that the iron smelting furnaces that had been spread in Korea before iron production appeared in Japan were completely different from box-shaped furnaces (Taylor, 1989; Park and Rehren, 2011; Nam, 2015). China did not know box-shaped furnaces either.

    • A view of iron and steel making technology in the Yan region during the Warring States period and the Han dynasty: Scientific study of iron objects excavated from Dongheishan site, Hebei province, China

      2014, Journal of Archaeological Science
      Citation Excerpt :

      Moreover, it has become clear that cast iron technology and bloomery iron technology both existed in the Yan region during this period. Considering the unmistakable transmission from the Yan region to Korea of iron artefacts and of cast iron technology (Nelson, 1993: 183; Park and Rehren, 2011; Shiomi, 1982; Wang, 1997: 285–340; Yoon, 1984; Yoon, 1986: 68–75; Yoon, 1989: 92–99), it is reasonable to suppose that also the bloomery iron technology may have been transported to Northeast China, Korea and Japan following the same route. It is clear that additional excavation and analytical work are called for on Yan-state iron objects, iron smelting sites and on other materials found in Northeast Asia in order to better understand the very real way that social and technological interactions in this area shaped the development of the Iron Age in East Asia.

    • Iron technology of the ancient megalithic communities in the Vidarbha region of India

      2013, Journal of Archaeological Science
      Citation Excerpt :

      These ancient artifacts were forged from low carbon iron as well and their fanned-ends were similarly carburized. The evidence presented for the practice of bloomery smelting in the southern part of the Korean peninsula (Park and Rehren, 2011) thus suggests that the Korean objects previously examined, like their Indian counterparts, were produced directly from low carbon bloomery iron rather than from the indirect and complicated processes involving cast iron. Granted, certain differences can be detected in the shape and extent of carburization in the objects examined from both regions; however, these minor discrepancies are not so severe as to deny the similarities and probable connections between them.

    • A preliminary study on the role and implication of plate-type iron artifacts in the ancient iron technology of Korea

      2012, Journal of Archaeological Science
      Citation Excerpt :

      Besides, the early form of plate-type objects is contemporaneous with or even predates the full establishment in China of such iron technology. By contrast, Park and Rehren (2011) found evidence of large-scale bloomery smelting practiced in the heartland of Korea in the AD 2nd to 3rd century. Their evidence confirms the theory proposed by Choi et al. (1998a,b) and Yun (1984), two pillars in Korean historical metallurgy, who did not discuss the technological and historical implications behind their theory at the time.

    View all citing articles on Scopus
    View full text