Next Article in Journal
Influence of Centerline Segregation Region on the Hydrogen Embrittlement Susceptibility of API 5L X80 Pipeline Steels
Previous Article in Journal
Laser Cutting of Titanium Alloy Plates: A Review of Processing, Microstructure, and Mechanical Properties
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

A Study on Metallurgical Artifacts Excavated from Luojiaba Site H235 in the Eastern Sichuan Region during the Eastern Han Dynasty

1
Institute of Cultural Heritage and History of Science and Technology, University of Science and Technology Beijing, Beijing 100083, China
2
Sichuan Provincial Institute of Archaeology and Cultural Relics, No. 5, Section 4, Renmin South Road, Chengdu 610041, China
*
Authors to whom correspondence should be addressed.
Metals 2024, 14(10), 1151; https://doi.org/10.3390/met14101151
Submission received: 11 September 2024 / Revised: 30 September 2024 / Accepted: 5 October 2024 / Published: 9 October 2024

Abstract

:
Currently, research remains limited on ironworking workshops in China and even throughout East Asia. The discovery of Luojiaba Site H235 in 2021 provides significant new material on this issue. This paper comprehensively organized the metallurgical artifacts and employed technological methods such as XRF and SEM-EDS to analyze the slag and hammerscale excavated from Luojiaba Site H235. First, the chemical composition and microstructural phases of the slag were analyzed, with an examination of the hammerscale. Second, the processes leading to the formation of slag and hammerscale are discussed. Third, based on the scale of the site, the chemical composition, and the characteristic shapes of the slag, combined with Serneels’ theoretical framework for quantitative analysis of forging slag, the slag unearthed from Luojiaba Site H235 was tentatively categorized as forging slag. This suggests that the workshop primarily produced or repaired iron objects through forging. Finally, considering the furnace shapes observed at contemporary sites and discussions on the potential sources of iron materials at the Luojiaba site, speculations are made regarding the furnace structure at the Luojiaba site and the possible origin of its iron materials. Unlike ironworking workshops at other contemporary sites, the location of the forging workshop at Luojiaba Site H235 presents a distinctly different choice, situated within a village or town during the Eastern Han Dynasty. This choice of location may reflect a phenomenon closer to consumer proximity. Therefore, analyzing metallurgical artifacts unearthed from Luojiaba contributes to understanding the operation of ironworking workshops and the distribution of resources in village-level settlements in the eastern Sichuan region during the Eastern Han Dynasty.

1. Introduction

The Iron Age in China began around the early 8th century BCE, during the Spring and Autumn period, marking the initial development of ancient iron metallurgy in China. During the subsequent Warring States period, iron metallurgy in China experienced rapid advancement, significantly enhancing the production of iron tools. By the Qin and Han dynasties, China had fully entered the Iron Age [1], characterized by the production of wrought iron. Even refined iron underwent decarburization and forging.
Currently, numerous iron artifacts produced through forging have been discovered throughout China and Central Asia. However, research on ironworking forges has been sparse and long overlooked by academia. The identification of ironworking forges and the reconstruction of the production chain for forged iron artifacts from that era are particularly lacking. The discovery of forging artifacts at the Luojiaba site (H235) filled a significant gap in this research. During the Han Dynasty, a critical period in the early development of iron smelting technology in China, extensive studies were conducted on smelting workshops in Henan [2,3,4], Shandong [5], Shaanxi [6], and other regions. These studies laid an important foundation for understanding Han-era iron production techniques, organization, and scale. In recent years, scientific and technological analyses of metallurgical remains from workshops have provided crucial materials for deeper research into Han Dynasty ironworking technology [7,8,9,10]. However, research on forging workshops remains scarce.
Sichuan, a significant iron-producing region during the Han Dynasty, has mostly seen analyses limited to furnace slag unearthed from iron smelting sites on the Chengdu Plain [11], as well as iron artifacts found in tombs and archaeological sites [12,13,14,15]. There has been relatively less analysis of metallurgical remains unearthed from Qin and Han period settlements or production workshops. The eastern Sichuan region has been particularly underexplored, with iron workshops discovered only at the Chengba (城坝) site, Quxian County, Sichuan Province [16]. After the Qin conquered the Ba and Shu regions, state-operated handicraft industries emerged in the area. In 311 BCE, alongside the construction of Chengdu city, Qin also established an iron office. During the Han Dynasty, four iron offices were established in the southwest: Linqiong (临邛) in Shu Commandery, Nan’an in Jianwei Commandery, Wuyang (武阳), and Mianyang (沔阳) in Hanzhong Commandery [17]. In the Eastern Han, an additional iron office was established in Dangqu (宕渠) in Ba Commandery [18]. Currently, systematic investigations or excavations have been conducted at three Han-era iron smelting sites on the Chengdu Plain: Gushishan (古石山) in Pujiang (蒲江), Tieniucun (铁牛村), and Xuxiebian (许鞋匾) [19,20,21] (Figure 1).
The Luojiaba site is located in Jinhua Village, Puguang Town, Xuanhan County, Dazhou City, Sichuan Province (Figure 1). The Sichuan Provincial Institute of Cultural Relics and Archaeology, along with other units, conducted archaeological excavations at the site in 1999, 2003, 2007, 2016, and from 2019 to 2022, totaling eight times. The cumulative excavated area is 5500 square meters. Over 200 tombs, more than 100 ash pits and the foundations of five houses were cleared. Excavated artifacts include over 3000 items of various types [22]. The metallurgical remains (H235) reported in this study were excavated in 2021. H235 measures 550 cm in length, 230 cm in width, and has a maximum depth of 80 cm. Excavated metallurgical artifacts include slag, remnants of iron objects, fragments of furnace walls, and charcoal. Household artifacts such as pottery shards, stone net weights, and animal bones were also found (Figure 2a,b). Based on the shape and decorative features of the pottery shards, archaeologists dated this deposit to the Eastern Han Dynasty (Figure 2c,d). Apart from H235, no furnaces, forging hearths, or smelting furnaces have been discovered. However, based on the unearthed forging debris and spherical forging artifacts, it is believed that iron forging activities occurred at this workshop. This study focuses on three main aspects: (1) analyzing the chemical composition of sampled slag to reconstruct and interpret relevant metallurgical processes; (2) characterizing the nature and technological processes of the slag and forging artifacts excavated from the workshop; and (3) exploring the functional aspects of the slag and forging artifacts corresponding to the workshop.

2. Materials and Methods

2.1. Sample Description

2.1.1. Slag

Using a strong magnet to assess magnetic properties and observing external characteristics, the slag unearthed from Luojiaba H235 can be macroscopically classified into two types: Type A exhibits strong magnetism, uneven surface texture with prominent protrusions and depressions predominantly covered with 0.1–0.2 cm thick reddish-brown rust stains, poor fluidity, low vitrification, and numerous internal pores. The cross-section appears dark brown when cut open (Figure 3a). Type B displays weaker magnetism and irregularly shaped flakes with uneven surface texture. The cross-section can be divided into two layers: the first layer is dense with visible sand grains, appearing reddish-brown, while the second layer is dark brown, porous, and loose, showing signs of vitrification indicative of higher heating temperatures (Figure 3b).

2.1.2. Forging Artifacts

The forging artifacts (hammerscale) primarily consist of spherical forging artifacts and forging waste flakes recovered through heavy liquid separation from samples at H235, which is deeply magnetic. Both exhibit a surface coloration of dark brown. The diameter of the spherical forging artifacts typically ranges from 0.5 to 0.6 cm, with 9 samples analyzed and numbered LJBH01 to LJBH09 (Figure 3c). The forging waste flakes have diameters mostly ranging from 0.3 to 0.5 cm, with 2 samples analyzed and numbered LJBH10 and LJBH11 (Figure 3d).

2.2. Experimental Methods

XRF Analysis: The samples were ground into powder using an oscillating mill and sieved. A total of 5 grams of sample powder were taken, backed with boric acid, and encapsulated to prepare the test specimens. Microchemical composition analysis of 31 samples was performed using a Bruker PUMA S2 energy dispersive X-ray fluorescence spectrometer (XRF) (Bruker Optics GmbH & Co., Ettlingen, Germany). The testing conditions included an Ag target, tube voltage of 40–50 kV, current of 2 mA, spot diameter of 28 mm, and SMART-Oxides mode. After testing, data were normalized to obtain accurate analysis results of oxide content in mass fractions. Results below the detection limit are indicated as “bdl” (below detection limit) in the data table.
Metallographic Analysis: Analysis was conducted on 32 samples excavated from Luojiaba Site H235. After numbering and photographing the slag relics, samples approximately 15–25 mm wide and 10–15 mm high were cut using a diamond band saw. Samples were embedded in epoxy resin using a metallographic mounter, initially polished with 600-grit sandpaper, and engraved with sample numbers. Subsequent polishing was performed using sandpaper and polishing cloth on a polishing machine until scratches were no longer visible under a Leica DM4000M metallographic microscope (Leica Camera AG, Wetzlar, Germany). Samples were carbon-coated after documenting critical positions on the sample surface.
Scanning Electron Microscope (SEM) Spectral Analysis: Using a TESCAN VEGA3 XMU scanning electron microscope (TESCAN, Brno, Czech Republic) equipped with a XFlash Detector 610 M spectrometer (Bruker Nano GmbH, Berlin, Germany), analysis was conducted on the matrix, metal particles, and other important phases of the samples. Imaging was performed in backscattered electron mode with an excitation voltage of 15 kV and working distance of 15 mm. Each spectral analysis lasted for 60 seconds. From 1 to 3 different areas of each sample were tested, and the average value was taken as the analysis result.

3. Results

3.1. Results of Slag Chemical Composition Analysis

The chemical composition of the slag is shown in Table 1. There are 22 Type-A slags, predominantly composed of Fe, Si, and Al, with minor amounts of Ca, Mg, Na, K, Ti, and other elements. The iron (Fe) content is highest, with FeO ranging from 58.1 to 86.7 wt%, SiO2 from 8.5 to 23.9 wt%, and Al2O3 from 2.6 to 8.5 wt%. Overall, Type-A slag is primarily characterized by the FeO–Al2O3–SiO2 system. There are 11 Type-B slags, primarily composed of Si, Al, and Fe, with minor amounts of Ca, Mn, Na, K, Ti, and other elements. The silicon (Si) content is highest, with SiO2 ranging from 54.3 to 66.3 wt%, FeO ranging from 8.6 to 20.3 wt%, and Al2O3 ranging from 12.0 to 17.8 wt%. Overall, Type-B slag is primarily characterized by the FeO–Al2O3–SiO2 system (Table 1).
No elements such as Cu, Pb, Zn, or Ag, nor their metal particles, were found in the slag. According to classification standards for ancient non-ferrous metal slag and iron smelting slag, the presence of copper (Cu), lead (Pb), zinc (Zn), or silver (Ag) slag can be excluded. Irregular iron particles were detected in the slag, which mainly consists of vitreous silicates.
Using principal component analysis (PCA) to reduce the dimensionality of slag composition data (Figure 4), further examination of differences between data sets was conducted. The results indicate that all of the 33 slag samples can be categorized into two types, with their compositions aligning well with the previously described slag classifications. Both types of slag primarily consist of FeO, SiO2, and Al2O3, but there are differences in the FeO and SiO2 content: Type A exhibits high FeO and low SiO2, while Type B shows low FeO and high SiO2. The Al2O3 content also varies between the two types: Type A averages 5.0 wt%, whereas Type B averages 15.0 wt%. Sample LJB15 in Type A has a significantly lower FeO content (58.1 wt%) compared to other samples in the same group.
Based on PCA, more groups were identified within Types A and B, leading to hierarchical clustering analysis (Figure 5). This analysis further subdivided Type A into 3 groups and Type B into 2 groups (Table 1). The main differences among the 3 groups in Type A are in the SiO2 and FeO content: Group A1 has the highest FeO content, averaging 84.7 wt%, followed by Group A2 at 72.0 wt%, and Group A3 with the lowest at 58.1 wt%. Conversely, SiO2 content shows the opposite trend: Group A1 has the lowest SiO2 content, averaging 10.1 wt%, Group A2 has 16.9 wt%, and Group A3 has the highest at 23.9 wt%. The distinction between the 2 groups in Type B also lies in SiO2 and FeO content: Group B1 averages 54.9 wt% SiO2 and 18.4 wt% FeO, while Group B2 averages 62.8 wt% SiO2 and 12.2 wt% FeO.

3.2. Microstructural Analysis Results

Microstructural analysis results show that in Type-A slag, Group A1, the predominant phases are wüstite and fayalite, with a glassy matrix and few or no visible iron particles. Wüstite occupies a large area in the A1 samples, with slightly different shapes; sample LJB01 is mostly elongated, while LJB25 is predominantly oval-shaped. Fayalite varies in shape; LJB01 appears as coarse elongated forms, whereas LJB25 shows finer elongated shapes (Figure 6a,b). Group A2 slag consists of wüstite, fayalite, glassy matrix, and iron particles. Wüstite is organized in oval shapes in some samples like LJB03 and LJB07, and in irregular arcs in others like LJB17. Fayalite appears in elongated forms, and iron particles exhibit irregular shapes (Figure 6c–e). Group A3 slag is composed mainly of fayalite and less wüstite, with fayalite distributed in elongated forms and less wüstite, with no visible iron particles (Figure 6f). In Type-B slag, Group B1 is primarily composed of silicate matrix with scattered crystallized Fe2O3 particles and magnetite in dendritic elongated forms. Hematite particles are irregularly distributed at the edges (Figure 7a,b). Group B2 is predominantly silicate matrix, with sparse zircon and wüstite particles. The wüstite particles exhibit irregular shapes and are locally densely distributed in needle-like and dendritic forms of magnetite. Magnetite and wüstite are assessed by shape and EDS oxygen (Figure 7c–f).
The glassy matrix of Type-A slag primarily consists of the following: Average 14.6 wt% Al2O3, average 14.0 wt% CaO, average 33.8 wt% FeO, and average 30.0 wt% SiO2. In Type-B slag, the glassy matrix primarily consists of the following: Average 16.2 wt% Al2O3, average 3.1 wt% CaO, average 32.2 wt% FeO, average 42.2 wt% SiO2 (Figure 8).
Hammerscale (forging scale) is typically attached to the surface of slag. For Type-A slag, hammerscale was found on the surfaces of LJB02, LJB05, and LJB19 (Figure 9a,b), while for Type-B slag, hammerscale was observed on the surfaces of LJB04, LJB28, LJB30, LJB32, LJB33, LJB34, and LJB36 (Figure 9c–f; Table 2), with Fe content ranging from 67.9 to 83.4 wt%.
Apart from being attached to the surfaces of slag, hammerscale was also found during gravity concentration of soil samples from H235. Additionally, spherical hammerscale artifacts were discovered (Figure 3c). The gravity concentration hammerscale exhibits chemical compositions consistent with those found in slag hammerscale, primarily FeO, with Fe contents of 76.0 wt% and 76.8 wt% (Figure 10a; Table 2). Inclusions were observed in the hammerscale (Figure 10b), and spherical hammerscale contained Fayalite, fayalite, and irregularly shaped iron particles (Figure 10c–f). The chemical compositions are predominantly Fe, Si, and Al, with small amounts of Ca, K, and P, where FeO ranges from 61.0 to 72.4 wt%, SiO2 ranges from 15.7 to 24.1 wt%, Al2O3 ranges from 3.4 to 7.6 wt%, and CaO ranges from 2.4 to 7.7 wt% (Table 3).

4. Discussion

4.1. The Chemical Composition and Classification of Slag

McDonnell suggested that the main components of smithing slag are oxides of Fe and Si, with minor oxides including Al2O3, K2O, and CaO [23,24,25]. The chemical composition of the slag from the Luojiaba site aligns with this. According to Serneels’ theoretical study on quantifying smithing slag [26], smithing slag is categorized into three types: SGD (scorie grise dense or dense gray slag), SAS (scorie arilo-sableuse or sandy clay slag), and SFR (scorie ferreuse rouillée or iron-rich slag).
The A1 group has the highest FeO content, averaging 84.7 wt%, with iron present as metallic and oxide particles, along with charcoal inclusions, and contains fayalite, classifying it as SFR, iron-rich slag. The SFR slags are generally rich and very rich in iron oxides and hydroxides (Figure 6a,b). The A2 group consists of Fayalite, irregular iron particles, and a minor glassy matrix, appearing gray, similar to slag from bloomery iron production, with hammerscale adhered to the surface, classifying it as SGD, dense gray slag (Figure 6c–e and Figure 9a,b). The B1 and B2 groups consist primarily of elements from sandstone and clay, with low iron content, abundant quartz, and colors ranging from black to brown to beige, containing identifiable remnants of unmelted sand grains, ceramic particles, and iron oxide skins, classifying them as SAS, sandy clay slag (Figure 7 and Figure 9c–f). The A3 group likely belongs to SGD-SAS, containing 0–10% sandy clay materials, 70–90% dense gray materials, and 10–20% iron-rich materials. With low FeO is probably a mix slag mainly SGD with some amount of SAS attached. Thus, compared to the other two groups in class A, A3 shows lower Fe and higher Si content (Figure 6f).
The shape (ellipticity, flatness, contour), appearance (color, porosity), physical properties (magnetism, apparent density), and characteristics (mineralogy and chemical composition) of smithing slag are interconnected to some extent. For instance, smithing slag entirely composed of SFR will be dense, covered with iron rust, and magnetic, often round-shaped, as seen in sample LJB25 of Group A1. Typical SAS slag will vary in color, have irregular shapes, be less compact, and have low apparent density, characteristics that all Luojiaba site Group B slag samples meet. Typical SGD slag will be flat and convex, have regular shapes, display a metallic gray fracture surface, have fewer pores, and show a dark gray and wine-red upper surface and a worm-like gray-rust lower surface, exemplified by samples LJB20, LJB23, LJB24, LJB26, etc., from Group A2. Additionally, during the re-floating of the H235 soil samples, numerous hammerscale fragments and spherical hammerscale artifacts were found (Figure 3c and Figure 9c–f), preliminarily indicating that the two types of slag unearthed from Luojiaba Site H235 are smithing slag.

4.2. Formation of Forging Slag and Debris

The most common type of slag related to forging activities is known as plano-convex smithing slag (PCB). This type of slag is typically associated with iron offcuts, various types of hammerscale (flat, irregular, spherical, etc.), nodular or irregularly shaped slag blocks, and furnace wall fragments. Plano-convex-shaped slag has been a common archaeological material across Europe from the Hallstatt period to modern times. As early as the 1960s, it was recognized as a type of metallurgical waste related to blacksmith shops. Archaeological and anthropological analyses have further substantiated this interpretation [27]. This type of slag is formed by the accumulation of molten materials at the bottom of the hearth [20,21,26] (Figure 11). The materials are introduced into the hearth where they come into contact with fire just above the tuyere, the highest temperature area. The accumulated materials melt, and the liquid slag flows downward and solidifies as it reaches lower temperatures. The A-type slag unearthed from Luojiaba Site H235 is a typical plano-convex slag formed in this manner. The following discusses how each element in A-type slag is introduced.
FeO is primarily introduced through thermal oxidation, the most significant process in the formation of forging slag. When oxygen comes into contact with iron, the oxidation process initiates. With continued heating, oxides form a hardened shell on the metal surface. Since the volume of oxides is greater than that of metal, the hardened shell eventually cracks with increasing volume, causing particles to fall into the hearth, facilitating slag formation. At lower temperatures and shorter durations, iron loss remains minimal. However, prolonged exposure and intense heat lead to significant losses.
SiO2, Al2O3, and other oxides originate from clay, sand, or stones and can be introduced in several ways. Examination of the melted surfaces on furnace lining fragments reveals localized melting due to high temperatures during forging. Craftsmen might introduce silica-rich sand into the furnace, or dust and other particles may inadvertently fall into the furnace from the working environment. Similar substances may be inadvertently introduced along with fuel if mixed during preparation or storage of the charcoal. In addition to these scenarios, blacksmiths commonly use sand or clay as fluxes or protective agents to cover the metal surface to prevent thermal oxidation. These substances create a melt on contact with iron oxide, protecting the metal from oxidation and decarburization at high temperatures. To achieve high-quality welding, it is crucial to avoid the formation of oxides at the joints.
Another significant source of materials is fuel. Traditional fuel for blacksmiths is charcoal, which varies widely in mineral content. Ash primarily contains lime (CaO) and potassium carbonate (K2O), along with small amounts of other substances (SiO2, MgO, P2O5, etc.). The morphology and size of some of the slag samples from the A-type slag unearthed at Luojiaba Site H235 indicate they belong to a single type: mostly circular with a concave–convex shape, with diameters reaching up to 10 cm. Rust adheres to the top, while burnt clay adheres to the bottom (Figure 11). These morphological features constitute typical examples of smithing hearth bottom (SHB) or plano-convex bottom (PCB), where each specimen likely represents a single forging operation [26]. Similar slag has been found at Tel Beth-Shemesh in Israel [28] and at the medieval harbor at Hoeke (Belgium) [29]. The AuŠra Selskienè study discusses chemical and phase composition and the microstructure of iron smelting and smithing slags from the Old Iron Age. The results of chemical analysis revealed that the smithing slags contained lower concentrations of phosphorus and manganese but a slightly larger amount of potassium as compared to the smelting [30]. In addition, there has been a great deal of discussion about this among Serneels, Perret, and other members of staff [31].
The Luojiaba site’s Class B slag exhibits higher levels of Si, Al, K, and P compared to Class A slag. Si, Al, and Ti likely originate from refractory materials and fluxing agents introduced during the smelting process, while K and P are likely derived from charcoal. Class B slag is lighter and forms on the periphery of the forge, with less contact with iron raw materials, resulting in its characteristic low Fe content.
Hammerscale, which includes iron oxide particles (flat or round), forms on the surface of metals during heating processes [32,33]. These particles detach from the metal, fall into the forge, and contribute to the formation of smithing slag. Additionally, during forging, hammerscale breaks into tiny particles and fragments on the anvil and in its vicinity [34]. Ironworkers add sand to the surface, creating a liquid that can be easily expelled through compression during forging at high temperatures. This process forms spherical hammerscale artifacts. The spherical hammerscale artifacts unearthed at the Luojiaba site were formed in a similar manner to those found in forging workshops in Kent County, England [35].

4.3. Smithing Hearth and Iron Raw Materials

At the Luojiaba site, no forge furnaces were found. However, three forge furnaces dating from the mid to late Eastern Han Dynasty to the Wei-Jin period were unearthed at the Chengba site in Quxian County (Figure 1), Sichuan Province, excavated by Sichuan University and the Sichuan Provincial Institute of Cultural Relics and Archaeology [16].
Forge furnace 1: This furnace had a rectangular plan with straight walls and a bottom lined with overlapping tile fragments. It measured 1.66 m in length, 0.68 m in width, and 0.2–0.24 m in height. The north and south walls were built with vertically arranged furnace bricks measuring approximately 0.32–0.42 m in length, 0.22 m in width, and 0.12–0.16 m in thickness, while the west wall was laid with half bricks measuring approximately 0.26 m in length, 0.2 m in width, and 0.12–0.16 m in thickness. All bricks used were reddish-brown refractory bricks with a thin layer of grey fine sand adhering to the inner walls. Only four bricks remained on the north wall, two on the west wall, and two on the south wall (Figure 12a).
Forge furnace 2: This furnace also had a rectangular plan with straight walls and a relatively flat bottom. It measured 0.52 m in length, 0.5 m in width, and 0.2 m in height. The north, south, and east walls were each built with a single refractory brick measuring approximately 0.32–0.36 m in length, 0.2 m in width, and 0.14 m in thickness, with a thin layer of grey fine sand adhering to the inner walls near the bottom and scattered tile fragments near the bottom (Figure 12b).
Forge furnace 3: This furnace had a rectangular plan with straight walls and a flat bottom. It measured 0.52 m in length, 0.44 m in width, and 0.22 m in height. The north, south, and east walls were built with a single rectangular brick measuring 0.41 m in length, 0.22 m in width, and 0.12 m in thickness, with circular and diamond-shaped decorations on the surfaces of the south and east walls (Figure 12c). Additionally, a large amount of slag and hammerscale was found nearby at G85 and G93 (G85 and G93 refer to the ash trench), indicating close association with smelting and forging activities.
Similar types of forge furnaces have also been found in the casting area of the Dongpingling City Site in Jinan, Shandong Province (L3) [36], the Tieshenggou Furnace in Gongyi, Henan Province [2], and the Han Dynasty forge found at Wafangzhuang, Beiguan, Nanyang [4]. According to The Chronicles of Huayang and The Book of Han, during the Western Han period, the production of ironware was primarily state-operated, with 49 government iron offices established nationwide. In the Eastern Han period, an additional iron office was set up in Dangqu, Ba Commandery (present-day Qu County, Sichuan) [18] (Figure 1).
The discovery of smelting remains, large building foundations, and over a hundred iron artifacts at the Chengba site in Quxian County confirms the production and use of iron tools on a significant scale. This evidence corroborates historical records of Dangqu having iron and Dangqu having an iron official [16]. The Luojiaba site, administratively under the jurisdiction of the Chengba site in Quxian County at the time, likely had its forging technology influenced by the Chengba site. The smithing hearth bottoms (SHB) at the Luojiaba site are circular, suggesting that the smithing hearths might have been circular as well, in contrast to the rectangular smithing hearths at the Chengba site. This indicates that the forging technology at Luojiaba Site H235 may have had broader influences. The iron raw materials at the Luojiaba site likely came from the Chengba site, but it is also possible that, within the context of a unified state, the raw materials came from the more distant Chengdu Plain or Nanyang in Henan [1].

5. Conclusions

Based on the analysis of metallurgical artifacts excavated from Luojiaba Site H235, it can be confirmed that the workshop represents a secondary stage of iron production. This stage involved forging iron billets into shapes or processing iron artifacts. The slag unearthed from H235 is identified as smithing slag based on its characteristic shapes, archaeological context, and the presence of associated hammerscale. Chemical and phase analyses support these interpretations, highlighting the difficulty in determining slag nature without context from the site itself. Based on slag bottom analysis, the forges at Luojiaba Site H235 likely had circular shapes.
The positioning of the Luojiaba site’s forge represents a distinct choice compared to other confirmed Han Dynasty forge discoveries such as Chengba in Qu County, Taicheng in Shaanxi, Dongpinglingcheng in Jinan, Tieshengou in Gongyi, Henan, and Wafangzhuang in Beiguan, Nanyang. The forges at Luojiaba Site H235 were smaller in scale and more specialized in function. This site’s location possibly reflects a phenomenon closer to consumer demand. Iron raw materials for the Luojiaba forges likely originated from nearby Chengba, where extensive iron smelting remains have been excavated in recent years. Chengba was also a documented center for iron production and administration during the Eastern Han Dynasty. However, it cannot be ruled out that iron materials came from farther locations such as the Chengdu Plain in Sichuan or Nanyang in Henan, facilitated by convenient waterway transportation. Further research is needed to explore these questions in greater detail.

Author Contributions

Conceptualization, G.L. and K.C.; methodology, C.L. and Y.L.; software, G.L.; validation, G.L., K.C., Y.L. and L.Z.; formal analysis, G.L. and L.Z.; investigation, W.C.; resources, L.Z.; data curation, G.L.; writing—original draft, G.L.; writing—review and editing, G.L.; visualization, G.L.; supervision, L.Z.; project administration, C.L.; funding acquisition, C.L. All authors have read and agreed to the published version of the manuscript.

Funding

This research was supported by the National Natural Science Foundation of China, Research on the Formation Mechanism of Non-Inclusions in Ancient Iron and Steel Products (52374296), a Sichuan Provincial Philosophy and Social Science Foundation Major Project, and Luojiaba Warring States Witch Tomb Collation and Research (SCJJ24ZD93).

Data Availability Statement

No new data were created or analyzed in this study. Data sharing is not applicable to this article.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Bai, Y.X. Archaeological Study of Iron Artifacts in Pre-Qin and Han Dynasties; Science Press: Beijing, China, 2005; pp. 48, 148, 351. (In Chinese) [Google Scholar]
  2. Cultural Relics Work Team of Henan Provincial Cultural Bureau. Iron Production Site at Tienshengou, Gongxian County; Cultural Relics Press: Beijing, China, 1962. (In Chinese) [Google Scholar]
  3. Zhengzhou Museum. Excavation Report of the Han Dynasty Iron Smelting Site at Guxing Town, Zhengzhou. Cult. Relics 1978, 2, 28–43. (In Chinese) [Google Scholar]
  4. Li, J.H. Excavation Report of the Han Dynasty Iron Smelting Site at Beiguan Wafangzhuang, Nanyang. Huaxia Archaeol. 1991, 1, 1–110. (In Chinese) [Google Scholar]
  5. Yang, Y.; Wei, C.M.; Xu, L.G. Archaeological Investigation and Excavation Achievements of Metallurgical Remains at the Qigucheng Site in Linzi, Shandong. Chinese Cultural Relics News, 19 July 2013; p. 8. (In Chinese) [Google Scholar]
  6. Shaanxi Provincial Institute of Archaeology. Taicheng Ironworks: Report on the Excavation and Research of a Cast Iron Foundry of the Han Period in Yangling, Shaanxi; Shanghai Classics Publishing House: Shanghai, China, 2018. (In Chinese) [Google Scholar]
  7. Chen, J.L. A New Exploration of Ancient Chinese Metal Metallurgy Civilization; Science Press: Beijing, China, 2014. (In Chinese) [Google Scholar]
  8. Du, N.; Li, J.X.; Zhang, G.M.; Wang, X.L.; Li, Y.X. Investigation and Study of the Iron Smelting Site in the Northeast Part of Qi Guo Ancient City, Linzi, Shandong. J. Jiangxi Univ. Sci. Technol. 2011, 6, 12–15. (In Chinese) [Google Scholar]
  9. Lin, Y.C.; Chen, J.L.; Zhong, J.R.; Lei, X.S.; Zhao, Y.P.; Chen, G. Sources and Circulation of Iron Production Materials in the Han Dynasty Guanzhong Region: Metallurgical Analysis of Iron Artifacts Excavated from the Taicheng Iron Foundry. Archaeol. Cult. Relics 2016, 6, 95–125. (In Chinese) [Google Scholar]
  10. Mo, L.H.; Lin, Y.C.; Chen, J.L. Excavation Report and Metallurgical Analysis of the Official Field Smelting Site in Sangzhi, Hunan. Cult. Relics South. China 2019, 3, 69–92. (In Chinese) [Google Scholar]
  11. Chen, J.L.; Yang, Y.D.; Zhou, Z.Q.; Chun, S. Preliminary analysis of smelting relics unearthed from the iron smelting site in Tieniu Village, Pujiang, Sichuan. Chendu Archaeol. Discov. 2009, 260–270. (In Chinese) [Google Scholar]
  12. Li, Y.; Yang, S.; Ma, C.; Yu, J. Research on Issues Related to Han Dynasty Iron Bridge Piers in Shitingjiang, Guanghan, Sichuan. Archaeology 2015, 9, 101–113. (In Chinese) [Google Scholar]
  13. Li, Y.; Zhang, M.; Zhou, K.; Yuan, H.; Li, Y. Preliminary Study on Iron Artifacts Excavated from Qiaogoutou Site in Yibin, Sichuan. South. Ethnol. Archaeol. 2021, 20, 417–438. (In Chinese) [Google Scholar]
  14. Zhang, M.; Ge, L.; Qiu, T.; Gan, Q.; Yang, B.; Li, Y. Forging or Casting: New Evidence of Iron Production in the Chengdu Plain in the Han Dynasty. Archaeol. Anthropol. Sci. 2021, 13, 111. [Google Scholar] [CrossRef]
  15. Li, Y.; Li, Y.; Yuan, H.; Juleff, G.; Zhang, M. Early Iron Objects of Southwest China: A Case Study of Iron Objects Excavated from Qiaogoutou Cemetery Site, Sichuan Province. Archaeol. Anthropol. Sci. 2019, 11, 1187–1198. [Google Scholar] [CrossRef]
  16. Chen, W.D.; Zhang, B.Y.; Liu, F.; Li, Y.F. Preliminary Report of Smelting Remains in 2021 Excavation at the Chengba Site, Quxian County, Sichuan Province. South Ethnol. Archaeol. 2023, 26, 1–31, (In Chinese with English Abstact). [Google Scholar]
  17. Ban, G. Geography Section. In Han Shu; Zhonghua Book Company: Beijing, China, 1962; Volume 28, pp. 1598, 1699. (In Chinese) [Google Scholar]
  18. Chang, Q.; Liu, L. Annotations to Chronicles of Huayang. In Ba Zhi · Dangqu Prefecture; Bashu Press: Chengdu, China, 1984; Volume 1, p. 99. (In Chinese) [Google Scholar]
  19. Li, Y.; Xiao, H.; Ma, C.; Murakami, Y.; Sun, T.; Li, Y. Microstructural and elemental analyses of slags excavated from the Gushishan iron-smelting site, Sichuan Province, China. Archaeometry 2024, 61, 1353–1365. [Google Scholar] [CrossRef]
  20. Li, Y.; Ma, C.; Murakami, Y.; Yang, Y.; Li, Y. Cast iron smleting and fining: An iron smelting site of the Eastern Han Dynasty in Xuxiebian, Sichuan Province, China. Sungkyun J. East Asian Stud. 2019, 19, 91–111. [Google Scholar]
  21. Yang, W.C.; Chen, Y.F. Preliminary Report on the Excavation of Iron-smelting Sites in Pujiang, Sichuan in 2007. Sichuan Cult. Relics 2008, 4, 17–26. (In Chinese) [Google Scholar]
  22. Sichuan Provincial Institute of Cultural Relics and Archaeology; Dazhou City Cultural Relics Management Office. Luojiaba Site in Xuanhan, Sichuan; Cultural Relics Press: Beijing, China, 2015. (In Chinese) [Google Scholar]
  23. Mcdonnell, G. The Classification of Early Ironworking Slags. Ph.D. Thesis, Aston University, Birmingham, UK, 1986. Volume 2. pp. 235 & 277. [Google Scholar]
  24. Mcdonnell, G. The study of early iron Smithing Residues. In The Craft of the Blacksmith, Symposium in Belfast; Scott, B., Cleere, H., Eds.; Symposium CSA: Belfast, UK, 1986; pp. 47–52. [Google Scholar]
  25. Mcdonnell, G. A model for the formation of smithing slags. Mater. Archeol. 1991, 26, 23–26. [Google Scholar]
  26. Serneels, V.; Perret, S. Quantification of smithing activities based on the investigation of slag and other material remains. In Archaeometallurgy in Europe II; Associazione Italiana di Metallurgia: Milano, Italy, 2003; pp. 469–478. [Google Scholar]
  27. Serneels, V. Archéométrie des scories de fer. In Recherches sur la Sidérurgie Ancienne en Suisse Occidentale; Cahiers D’archéologie Romande: Lausanne, Switzerland, 1993; Volume 61. [Google Scholar]
  28. Veldhuijzen, H.A.; Rehren, T. Slags and the City: Early Iron Production at Tell Hammeh, Jordan, and Tel Beth-Shemesh, Israel. In Metails and Mines-Studies in Archaeometallurgy; Archetype; British Museum: London, UK, 2006; pp. 189–201. [Google Scholar]
  29. Baulina, P.; Clercq, W.; Dewaele, S.; Vanhaecke, F.; Grave, J. Archaeometallurgical research into the ironworking activities of the Medieval Harbour at Hoeke (Belgium). STAR Sci. Technol. Archaeol. Res. 2023, 9, 1–13. [Google Scholar]
  30. Selskienè, A. Examination of smelting and smithing slags formed in bloomery iron-making process. CHEMIJA 2007, 18, 22–28. [Google Scholar]
  31. Serneels, V.; Duvauchelle, A. Des Artisans à la Campagne. Carrière de Meules, Forge et Voie Gallo-Romaine à Châbles (FR); Archéologie Fribourgeoise; Academic Press Fribourg: Fribourg, Switzerland, 2003; Volume 19, pp. 77–174. [Google Scholar]
  32. Dungworth, D.; Wilkes, R. Understanding Hammerscale: The Use of High-Speed Film and Electron Microscopy. Hist. Metall. 2009, 43, 33–46. [Google Scholar]
  33. Young, T. Some Preliminary Observations on Hammerscale and Its Implications for Understanding Welding. Hist. Metall. 2011, 45, 26–41. [Google Scholar]
  34. Soulignac, R.; Serneels, V. Forging with Dogon Smiths (Mali). In Accidental and Experimental Archaeometallurgy; Historical Metallurgical Society: London, UK, 2012; pp. 119–126. [Google Scholar]
  35. Paynter, S. Romano-British Workshops for Iron Smelting and Smithing at Westhawk Farm, Kent. Hist. Metall. 2007, 41, 15–31. [Google Scholar]
  36. Zhang, S.; Zhao, H.C.; Zheng, T.X.; Yang, Z.F.; Chen, J.L.; Wang, Z.B.; Zhang, Z.G. Excavation Report on the Foundry Area of Dongpinglingcheng Site in Zhangqiu District, Jinan City, 2009. Archaeology 2019, 11, 49–66. (In Chinese) [Google Scholar]
Figure 1. Geographical location schematic of Luojiaba site in Sichuan Province, China.
Figure 1. Geographical location schematic of Luojiaba site in Sichuan Province, China.
Metals 14 01151 g001
Figure 2. Schematic map of Luojiaba site in Xuanhan County, Sichuan Province, and related information of H235. (a) Location map of Luojiaba site; (b) photo of Luojiaba Site H235 (refer to the official website of Sichuan Provincial Institute of Cultural Relics and Archaeology); (c,d) pottery shards unearthed from Luojiaba Site H235.
Figure 2. Schematic map of Luojiaba site in Xuanhan County, Sichuan Province, and related information of H235. (a) Location map of Luojiaba site; (b) photo of Luojiaba Site H235 (refer to the official website of Sichuan Provincial Institute of Cultural Relics and Archaeology); (c,d) pottery shards unearthed from Luojiaba Site H235.
Metals 14 01151 g002
Figure 3. Artifacts unearthed from Luojiaba Site H235. (a) Type-A slag; (b) type-B slag; (c) spherical forging artifacts; (d) forging waste flakes.
Figure 3. Artifacts unearthed from Luojiaba Site H235. (a) Type-A slag; (b) type-B slag; (c) spherical forging artifacts; (d) forging waste flakes.
Metals 14 01151 g003
Figure 4. Principal component analysis (PCA) and scatter plots of chemical composition at Luojiaba site. (a) Eigenvector plot and PCA scatter plot; (b) FeO–SiO2 scatter plot of chemical composition; (c) Al2O3–SiO2 scatter plot of chemical composition; (d) CaO–MgO scatter plot of chemical composition.
Figure 4. Principal component analysis (PCA) and scatter plots of chemical composition at Luojiaba site. (a) Eigenvector plot and PCA scatter plot; (b) FeO–SiO2 scatter plot of chemical composition; (c) Al2O3–SiO2 scatter plot of chemical composition; (d) CaO–MgO scatter plot of chemical composition.
Metals 14 01151 g004
Figure 5. Hierarchical clustering analysis of slag chemical composition.
Figure 5. Hierarchical clustering analysis of slag chemical composition.
Metals 14 01151 g005
Figure 6. Scanning electron microscope backscatter electron images of Type-A slag from Luojiaba site. Wü: Wüstite (FeO); Fa: Fayalite; Gl: Glassy matrix; Fe: Iron particles. Group A1: (a) LJB01; (b) LJB25; Group A2: (c) LJB03; (d) LJB07; (e) LJB17; Group A3: (f) LJB15.
Figure 6. Scanning electron microscope backscatter electron images of Type-A slag from Luojiaba site. Wü: Wüstite (FeO); Fa: Fayalite; Gl: Glassy matrix; Fe: Iron particles. Group A1: (a) LJB01; (b) LJB25; Group A2: (c) LJB03; (d) LJB07; (e) LJB17; Group A3: (f) LJB15.
Metals 14 01151 g006
Figure 7. Backscattered electron microscope images of Type-B Slag from Luojiaba site. Wü: Wüstite; ZrSiO4: Zirconium silicate (Zircon); Gl: Glassy matrix; Mag: Magnetite. Group B1: (a) LJB04; (b) LJB37; Group B2: (c) LJB28; (d) LJB30; (e) LJB31; (f) LJB34.
Figure 7. Backscattered electron microscope images of Type-B Slag from Luojiaba site. Wü: Wüstite; ZrSiO4: Zirconium silicate (Zircon); Gl: Glassy matrix; Mag: Magnetite. Group B1: (a) LJB04; (b) LJB37; Group B2: (c) LJB28; (d) LJB30; (e) LJB31; (f) LJB34.
Metals 14 01151 g007
Figure 8. Box plots of glassy matrix composition percentages for two types of slag at Luojiaba site.
Figure 8. Box plots of glassy matrix composition percentages for two types of slag at Luojiaba site.
Metals 14 01151 g008
Figure 9. Backscattered electron microscope images of hammerscale attached to slag surfaces. Type A: (a) LJB02, (b) LJB19; Type B: (c) LJB04, (d) LJB28, (e) LJB30, (f) LJB33.
Figure 9. Backscattered electron microscope images of hammerscale attached to slag surfaces. Type A: (a) LJB02, (b) LJB19; Type B: (c) LJB04, (d) LJB28, (e) LJB30, (f) LJB33.
Metals 14 01151 g009
Figure 10. Backscattered electron microscope images of hammerscale. (a) LJBH11 hammerscale fragment; (b) inclusions in hammerscale fragment LJBH11; (c) LJBH05 spherical hammerscale artifact; (d) close-up of LJBH05 spherical hammerscale artifact: light gray phase represents fayalite, gray phase represents Fayalite, black phase represents glassy phase, dark black phase represents holes; (e) LJBH09 spherical hammerscale artifact; (f) partial view of LJBH09 spherical hammerscale artifact: bright phase represents iron particles, light gray phase represents fayalite, gray phase represents Fayalite, black phase represents glassy phase, dark black phase represents voids.
Figure 10. Backscattered electron microscope images of hammerscale. (a) LJBH11 hammerscale fragment; (b) inclusions in hammerscale fragment LJBH11; (c) LJBH05 spherical hammerscale artifact; (d) close-up of LJBH05 spherical hammerscale artifact: light gray phase represents fayalite, gray phase represents Fayalite, black phase represents glassy phase, dark black phase represents holes; (e) LJBH09 spherical hammerscale artifact; (f) partial view of LJBH09 spherical hammerscale artifact: bright phase represents iron particles, light gray phase represents fayalite, gray phase represents Fayalite, black phase represents glassy phase, dark black phase represents voids.
Metals 14 01151 g010
Figure 11. Luojiaba Site H235: Top and profile views of some smithing hearth bottom (SHB) slags. The SHB slags clearly show a convex–concave porous core, with rust attached to the top and embedded soil at the bottom. (a) LJB19, (b) LJB21, (c) LJB22, (d) LJB27.
Figure 11. Luojiaba Site H235: Top and profile views of some smithing hearth bottom (SHB) slags. The SHB slags clearly show a convex–concave porous core, with rust attached to the top and embedded soil at the bottom. (a) LJB19, (b) LJB21, (c) LJB22, (d) LJB27.
Metals 14 01151 g011
Figure 12. Cross-sectional view of forges at Chengba site [16]. (a) Forge furnace 1, (b) Forge furnace 2, (c) Forge furnace 3.
Figure 12. Cross-sectional view of forges at Chengba site [16]. (a) Forge furnace 1, (b) Forge furnace 2, (c) Forge furnace 3.
Metals 14 01151 g012
Table 1. Chemical composition of slag.
Table 1. Chemical composition of slag.
Sample IDMass Fraction (wt%)Group
Na2OMgOAl2O3SiO2P2O5K2OCaOTiO2MnOFeO
LJB010.10.62.911.80.40.41.10.1bdl82.7A1
LJB020.21.04.516.70.70.74.70.20.770.8A2
LJB03bdl0.64.513.60.90.51.30.21.177.2A2
LJB050.30.74.618.20.80.60.40.20.274.1A2
LJB060.40.95.719.60.30.82.00.30.769.2A2
LJB070.10.75.616.40.50.81.70.21.472.6A2
LJB08bdl0.75.915.90.30.90.90.21.573.6A2
LJB09bdl1.25.516.20.50.83.90.21.270.5A2
LJB10bdl0.75.114.80.40.71.40.21.175.5A2
LJB140.30.84.815.70.40.62.40.20.374.4A2
LJB150.41.58.523.91.21.23.70.41.258.1A3
LJB160.31.05.021.60.80.75.00.30.265.1A
LJB170.60.95.117.00.30.73.50.20.371.4A2
LJB190.61.04.516.60.50.84.10.10.171.5A2
LJB200.10.63.516.90.20.32.90.1bdl75.2A2
LJB210.31.04.518.30.60.75.60.20.368.5A2
LJB220.10.75.815.71.10.51.00.31.173.7A2
LJB230.20.84.615.50.80.63.00.21.073.3A2
LJB240.20.96.417.80.70.81.30.21.170.6A2
LJB25bdl0.52.68.50.50.30.6bdl0.386.7A1
LJB260.21.13.714.31.00.45.40.10.173.7A2
LJB270.31.07.419.90.60.71.00.31.267.7A2
LJB042.91.817.855.51.52.01.20.70.116.5B1
LJB283.92.016.362.80.42.22.10.70.19.5B2
LJB293.11.817.863.01.22.01.10.80.19.2B2
LJB303.21.716.664.61.12.11.30.80.18.6B2
LJB313.11.715.359.80.81.91.30.70.115.3B2
LJB323.81.915.360.30.42.12.50.70.113.0B2
LJB331.51.713.066.31.51.74.40.80.18.9B2
LJB341.41.712.460.81.71.63.00.70.116.6B2
LJB351.61.912.063.30.42.03.70.70.114.4B2
LJB361.81.812.964.21.01.81.50.70.114.0B2
LJB373.31.715.654.31.21.81.20.60.120.3B1
Table 2. Composition of hammerscale adhering to slags and gravity concentration hammerscale.
Table 2. Composition of hammerscale adhering to slags and gravity concentration hammerscale.
Sample IDMass Fraction (wt%)
Al2O3SiO2CaOFeO
Composition of hammerscale adhering to slagsLJB02-1bdlbdlbdl100.0
LJB02-2bdlbdlbdl100.0
LJB02-3bdl1.4bdl98.6
LJB05bdl0.4bdl99.6
LJB19bdl1.2bdl98.8
LJB04-10.60.7bdl98.7
LJB04-21.62.1bdl96.3
LJB28-1bdlbdlbdl100.0
LJB28-21.52.8bdl95.6
LJB30-12.36.4bdl91.4
LJB30-23.66.9bdl89.5
LJB32-11.01.1bdl97.9
LJB32-21.52.2bdl96.3
LJB33bdlbdlbdl100.0
LJB341.72.00.396.0
LJB362.23.50.394.0
Gravity concentration hammerscaleLJBH100.93.90.594.7
LJBH11bdl3.00.796.4
Table 3. Chemical composition of spherical forging remnants.
Table 3. Chemical composition of spherical forging remnants.
Sample IDMass Fraction (wt%)
Al2O3SiO2P2O5K2OCaOFeO
LJBH014.218.12.10.62.672.4
LJBH026.020.50.41.03.268.8
LJBH037.624.11.02.43.961.0
LJBH054.820.00.00.62.472.3
LJBH063.415.70.91.07.771.3
LJBH097.619.90.71.22.468.3
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Li, G.; Li, Y.; Li, C.; Chen, K.; Zheng, L.; Chen, W. A Study on Metallurgical Artifacts Excavated from Luojiaba Site H235 in the Eastern Sichuan Region during the Eastern Han Dynasty. Metals 2024, 14, 1151. https://doi.org/10.3390/met14101151

AMA Style

Li G, Li Y, Li C, Chen K, Zheng L, Chen W. A Study on Metallurgical Artifacts Excavated from Luojiaba Site H235 in the Eastern Sichuan Region during the Eastern Han Dynasty. Metals. 2024; 14(10):1151. https://doi.org/10.3390/met14101151

Chicago/Turabian Style

Li, Gaobo, Yanxiang Li, Chenyuan Li, Kunlong Chen, Luhong Zheng, and Weidong Chen. 2024. "A Study on Metallurgical Artifacts Excavated from Luojiaba Site H235 in the Eastern Sichuan Region during the Eastern Han Dynasty" Metals 14, no. 10: 1151. https://doi.org/10.3390/met14101151

APA Style

Li, G., Li, Y., Li, C., Chen, K., Zheng, L., & Chen, W. (2024). A Study on Metallurgical Artifacts Excavated from Luojiaba Site H235 in the Eastern Sichuan Region during the Eastern Han Dynasty. Metals, 14(10), 1151. https://doi.org/10.3390/met14101151

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop