Introduction

Pottery adorned with paints possesses a rich historical background, as demonstrated by global archeological records that reveal considerable diversity in both technological and artistic practices1,2,3,4,5,6. Decorated pottery can be categorized into pre-firing painted pottery (PRFP pottery彩陶) and post-firing painted pottery (POFP pottery彩绘陶), distinguished by different painting techniques utilized7,8,9. It is crucial to distinguish between PRFP and POFP pottery, as they exhibit unique painting techniques, utilize different types of paints, and possess distinct functional characteristics7. Currently, the painting processes of painted pottery have not been subjected to scientific investigation. The distinction between PRFP and POFP pottery primarily relies on visual analysis of the color and preservation state of the paints. Studies have demonstrated that the paint used in POFP pottery exhibits more vibrant colors, which are likely to retain their original hue. However, this type of paint is also more susceptible to flaking or complete detachment. In contrast, the paint utilized in PRFP pottery is solidified at high temperatures and tightly adheres to the ceramic body, making it less prone to chipping or falling off. Consequently, PRFP pottery is typically well-preserved when unearthed8,10,11.

The Late Neolithic Age ((circa 4500–4000 BC) marked the emergence of social complexity in China, characterized by increased exchanges and interactions among the upper classes across various regions. These interactions involved items such as jade, ivory artifacts, wide-mouth jars, and axes. Concurrently, the prevalence of POFP pottery gradually increased, with numerous examples uncovered at burial sites11. Research has shown that post-firing painting was intentionally selected over the more widely used pre-firing painting technique, which was prevalent among Late Neolithic communities in China, particularly in the lower Yellow River and Yangtze River basins11. The POFP pottery serves as a crucial medium for communication and dissemination among the upper classes across various regions, highlighting its significant role in reflecting social hierarchy10,11. Additionally, POFP pottery has been discovered at archeological sites associated with the Yangshao culture (approximately 4500–2700 BC) and the Longshan culture (approximately 2700–1900 BC), both of which are situated in the middle reaches of the Yellow River11. However, only a limited portion of the POFP pottery has undergone scientific analysis, primarily focusing on the detection of components in pottery surface paints [12, 13], the painting techniques and processes associated with POFP pottery from the Yangshao period have not been thoroughly investigated.

The Nanzuo site is located in Nanzuo Village (南佐村), within Houguanzhai Town (后官寨镇), Xifeng District (西峰区), Qingyang City (庆阳市), Gansu Province (甘肃省) (see Fig. 1). Discovered in 1957, the site underwent six archeological excavations from 1986 to 1996. Since 2021, new excavations have expanded the confirmed area to over 6 million square meters, with carbon-14 dating indicating an age of the remains of 5100–4700 years ago14. The Nanzuo site is composed of various functional areas (see Fig. 2), including the “Palace Area”, the “Core Area”, and the “Residence Areas.” Notably, large rammed earth buildings were found in the “Palace Area”, likely serving as the highest-ranking administrative and sacrificial spaces. There are also densely distributed buildings in the “Core Area”, which may have served as aristocratic residences, surrounded by ring trenches covering approximately 300,000 square meters. Additionally, the “Residence Areas” reveal numerous densely distributed cave dwellings. Overall, the Nanzuo site appears to have functioned as a large-scale, high-ranking central settlement, reminiscent of a capital city on the Loess Plateau, dating back to the late Yangshao culture. These findings suggest early stages of state formation or a developed society on the Loess Plateau (黄土高原) 5000 years ago. Notably, approximately one hundred sherds of POFP pottery were unearthed in the “Palace Area” (see Fig. 2).

Fig. 1: Distribution of POFP pottery sites unearthed in the middle reaches of the Yellow River.
figure 1

1. Xipo; 2. Xinjie; 3. Shizhaocun; 4. Dadiwan; 5. Nanzuo; 6. Shimao; 7. Zhaishan; 8. Xiajin; 9. Taosi; 10. Qingliangsi.

Full size image
Fig. 2: Excavation location of POFP pottery in the Nanzuo site.
figure 2

a Functional distribution of the Nanzuo site; b The POFP pottery sherds excavated from the Palace Area).

Full size image

In this paper, we conduct a comprehensive study of the paints and painting techniques employed in Nanzuo POFP pottery. Firstly, we observe both the macro characteristics and micro morphology of the painted sections of POFP pottery. Subsequently, we analyze the composition of the paints to reconstruct the painting process utilized in creating POFP pottery. Finally, we perform a comparative analysis of the painting techniques found in POFP pottery with those from other Neolithic sites.

Methods

Description of samples

To determine the painting materials and techniques of Nanzuo POFP pottery, this paper selected the five most representative pottery sherds for research. These samples were chosen for their excellent preservation of paints and their ability to represent all the POFP potteries found in the unearthed units. The five sherds in this paper were numbered NZ1 (2021QNF2③:2), NZ2 (2021QNF2③:3), NZ3 (2021QNF2③:1), NZ4 (collected), and NZ5 (96XNT402) (see Fig. 3). All samples were sherds of POFP potteries with residual red paints on the surface. Due to soil attachment in the buried environment, the surface of the NZ1 and NZ2 appears gray, but the paint layer remains intact. Four of the samples (NZ1, NZ2, NZ3, NZ5) should be classified as the same type of artifacts, with red paint on their external (convex) side and featuring protrusions with impressed pits on the surface (see Fig. 3). However, it is worth noting that NZ4 differs from the others, as it is painted on the internal (concave) side of the pottery’s bottom (see Fig. 3-NZ4). NZ1, NZ2, and NZ3 were discovered in unit F2.

Fig. 3: POFP pottery sherds unearthed at the Nanzuo site.
figure 3

(NZ1. 2021QNF2③:2; NZ2. 2021QNF2③:3; NZ3. 2021QNF2③:1; NZ4. collected, NZ5. 96XNT402).

Full size image

Microscopic observation

This paper presents a comprehensive study of POFP pottery sherds from Nanzuo. The research involved the utilization of an ultra-depth microscope and scanning electron microscopy (SEM) to examine the surface characteristics of the sherds, while cross-sectional analysis was conducted using both optical microscopy and SEM techniques. The ultra-depth microscope employed in this investigation was a KH-7700 model produced by the United States Coastal Company. Additionally, a ZEISS Primotech microscope was utilized in conjunction with a tablet computer and Matscope software to generate high-quality micrographs exhibiting exceptional clarity. The findings from the SEM and energy-dispersive spectroscopy (EDS) analyses are detailed in subsequent sections.

Chemical compositional analysis

In order to further investigate the decorative coating process of Nanzuo’s POFP pottery, we conducted an analysis of the elemental composition of the paints present on both the surface and cross-section of the pottery. This was achieved using a scanning electron microscope coupled with energy dispersive spectroscopy (SEM-EDS), specifically employing a German ZEISS EVO 25 scanning electron microscope alongside a British Oxford X-Max 20 energy spectrometer. The SEM-EDS system is equipped with a 20 mm² silicon drift detector (SDD), which provides an impressive energy resolution of 127 eV. The working distance was maintained at 8.5 mm, while the acquisition and counting rate for the energy spectrum reached up to 3 Kcps. Each spectrum acquisition lasted for 60 s, and during analysis, the dead time remained below 30%.

The phase of the paints was analyzed using laser microscopy in conjunction with a confocal Raman spectrometer (model In Via, Renishaw Co., Ltd.). The Raman spectroscopy was conducted under the following conditions: the excitation wavelength of the argon ion laser source was set at 532 nm, with a spectral range spanning from 200 to 1060 nm and a spectral resolution between 0.5 and 1 cm. Throughout the experiment, it was generally required that the temperature be maintained between 18 and 30 °C, while humidity levels were kept below 50%.

To ascertain whether a binding medium was utilized between the paints and the pottery, an analysis of the paints was conducted using Fourier Transform Infrared Spectroscopy (FT-IR). The infrared instrument employed for this analysis was the Nicolet In10 FT-IR Microscope, manufactured by Thermo Fisher Scientific in the United States, equipped with an MCT/A detector and a BaF2 window. Furthermore, to investigate whether the binding medium contained organic compounds, protein extraction and subsequent testing of NZ2 samples were undertaken.

Results

Paints composition

According to the SEM observations of all the peeling paints from the POFP potteries in this study, the results indicated a distinct color difference between the internal and external layers of the paints. The external layer exhibited a red hue, while the internal layer appeared black (see Fig. 4b). The surface of the red layer was uneven, characterized by a loose structure with no discernible burnishing traces (see Fig. 4a). Its thickness measured less than 10 micrometers (see Fig. 4d). In contrast, the black layer presented a flat profile with uniform particle size distribution (see Fig. 4c), featuring a relatively dense structure and an approximate thickness of 10 micrometers (see Fig. 4d).

Fig. 4: Micromorphology of the paints of NZ2.
figure 4

a The red external layer of the paints; b Paints’ chips; c Black internal layer of the paints; d SEM image of paints in cross-section; e Hg (elemental mercury) content of the paints in cross-section; f S (element sulfur) content of paints in cross-section).

Full size image

The SEM-EDS analysis (Table 1, Fig. 5a–c) and Raman spectroscopy (Fig. 5d: peak 250, peak 340) confirmed that the primary component of the red external layer is cinnabar (HgS)13,15. Sulfur and mercury were predominantly concentrated in the red external layer, while their concentrations in the black internal layer were significantly lower (see Fig. 4e, f). The SEM-EDS analysis further revealed that the black internal layer is primarily composed of carbon and oxygen compounds (Fig. 5, Table 1). Moreover, the levels of calcium, aluminum, and silicon in the black internal layer exceeded those found in the external layer (Table 1), suggesting potential diffusion of these elements between the internal layer and the pottery body. Previous studies have also indicated that under conditions of high humidity or intense light exposure, cinnabar can darken, implying that this paint was applied and treated after firing8,16 (see Fig. 4a).

Table 1 Elemental analysis of Nanzuo’s POFP pottery: Body and paints
Full size table
Fig. 5: Analytical test points from NZ2 and NZ5.
figure 5

a Cross-section of NZ5; b Cross-section of NZ5 with a flute; c Cross-section of paints of NZ2; d Raman result of red paints on NZ5, A–G test point in Table 1—the results of A–G are shown in Table 1.

Full size image

This study posits that the black internal layer serves as a binding medium, effectively securing the cinnabar paints to the pottery body. Furthermore, infrared analysis and protein extraction of the black layer revealed an absence of organic matter.

Painting process

In this study, we meticulously reconstructed the detailed processes involved in the application of paints on POFP pottery:

  1. (1)

    Roughening Pretreatment. The Nanzuo POFP pottery displays distinct scratches beneath the surface paints, which can be clearly observed through ultra-depth microscopy. When magnified 500 times, these scratches exhibit variations in depth, width, length, and direction, contributing to an overall disordered appearance. Notably, even on the internal side of the NZ4 specimen (collected), evident signs of roughening are present (see Fig. 3). This deliberate roughening process, conducted prior to the application of paints, appears to represent a distinctive and unique technical characteristic of Nanzuo’s POFP pottery. Empirical evidence suggests that paints applied to these roughened areas demonstrate enhanced preservation (see Fig. 6).

  2. (2)

    Application of the binding medium. In this phase, a cross-section of the red paints with the pottery body in NZ5 was embedded in resin. The micromorphology of the cross-section was then observed under 500x magnification using an optical microscope. The red and black layers were visually depicted through cross-sectional photographs, allowing for an intuitive presentation. The binding medium effectively permeates the pottery’s scratched surface, enhancing the bonding of the paints to the pottery. The surface of the black binding medium is relatively flat, suggesting a potential polishing step following the application of the binding medium layer (Fig. 7).

  3. (3)

    Application of Cinnabar. Upon examining the cross-section, it was observed that the cinnabar layer is remarkably thin, indicative of the high level of craftsmanship exhibited by ancient artisans. Cinnabar primarily comprises sulfur and mercury, with relatively low concentrations of other mineral elements; some of these elements are present at levels below the detection threshold.

Fig. 6
figure 6

Micrograph of paints on pottery from Nanzuo.

Full size image
Fig. 7
figure 7

Micrograph of the cross-section of NZ5.

Full size image

Discussion

This research has demonstrated that the painting process of POFP pottery in Nanzuo is considerably more intricate than previously understood. The identification of roughening pretreatment, the application of binding mediums, and the incorporation of cinnabar represent significant advancements in our understanding of Neolithic painting techniques.

First and foremost, the surface polishing of pottery constituted a significant process during the Neolithic Age. A variety of methods were employed, including the utilization of fine mud as a raw material to achieve smoother surfaces. Furthermore, polishing frequently followed painting in order to enhance both the smoothness and esthetic appeal of the pottery7. Ethnological surveys reveal that techniques for pottery polishing continue to be practiced in contemporary workshops. However, in contrast to previous studies, it is noteworthy that the surface of Nanzuo’s POFP pottery was intentionally roughened before paint application. This technique was specifically designed to improve paint adhesion. Such a distinctive characteristic appears unique to the Nanzuo site and plays a crucial role in preserving paints on Nanzuo POFP pottery.

Secondly, the binding medium plays a crucial role in painted cultural relics, facilitating the adhesion of paints to their surfaces. Qingling Ma speculated that the painted layers at the Dadiwan site (circa 3100–2800 BC) may have utilized a binding medium; however, there is currently no scientific evidence to substantiate this claim12. Naisheng Li investigated the application of slaked lime (Ca(OH)2) as a binding medium between pottery surfaces and cinnabar paints at the Taosi site (circa 2300–1900 BC)13. The use of organic cement was prevalent in ancient murals, where animal glue typically served as the binding medium17,18.

At Nanzuo, black binding medium was applied to roughened pottery surfaces; this was confirmed by SEM-EDS analysis which identified it as a carbon-oxygen compound. Black pigments were frequently utilized in ancient pottery, figurines, and murals, with carbon black being the primary component13,17,18,19,20. Therefore, we hypothesize that the black pigments found at the Nanzuo site are also likely to be composed of carbon black. This instance represents one of China’s earliest known applications of binding mediums. This method may have influenced subsequent techniques used in POFP pottery production.

Finally, in comparison to iron red used in PRFP pottery decoration1,7,21,22,23, cinnabar exerts a more pronounced visual impact as a red pigment. Consequently, scholars both domestically and internationally generally associate the use of cinnabar with themes of death, religion, and ritual practices. For example, Neolithic burials and wall paintings discovered at Çatalhöyük in Turkey (circa 7500–6000 BC) were created using cinnabar paints, predominantly applied to interred males24. Cinnabar was also employed for decorating figurines during the Early Cycladic II period (circa 2400–2300 BC) in the Cyclades region of Greece; its exotic origins likely conferred additional prestige upon this color among the islanders25. At the Olmec site of La Venta (circa 900–400 BC), elite individuals and their grave goods were adorned with substantial amounts of cinnabar26. Furthermore, a ritual cache of cinnabar was uncovered beneath a ballcourt at Lamanai in Belize, dating back to the 10th century CE (Terminal Classic)27.

Cinnabar was extensively utilized in ancient China7,28,29,30,31,32. Its applications included the decoration of walls and floors in buildings specifically designated for ritual ceremonies, its incorporation into burial practices (notably in cinnabar-bottomed burials), its use as a pigment in lacquerware, and as red paint on POFP potteries. However, at the Nanzuo site, most of these POFP pottery sherds are believed to be fragments of ceramic drums, which were common high-grade musical instruments during the Neolithic period in China. This suggests that POFP potteries may have served a more diverse range of functions.

This paper presents a comprehensive study of the paint materials and painting techniques used in Nanzuo POFP pottery, utilizing Optical Microscopy (OM), X-Ray Fluorescence (XRF), Infrared Fourier Transform Spectroscopy (IFIR), and Scanning Electron Microscopy with Energy Dispersive X-ray Spectroscopy (SEM-EDS). The key findings are as follows:

First, The surface and cross-section of the POFP pottery were examined using scanning electron microscopy. It was observed that traces of roughening pretreatment were present on the pottery’s surface. The paint consisted of two distinct layers: the outer layer, composed of red cinnabar, exhibited a loose structure with an uneven surface; whereas the inner layer, made up of a black binding medium, displayed a flat and dense structure.

Second, The technique of roughening pretreatment is currently exclusive to the POFP pottery of Nanzuo. Unlike conventional practices that focus on polishing pottery surfaces, the POFP pottery from Nanzuo employs an intentional roughening process designed to enhance the properties of its binding medium. This distinctive approach plays a vital role in preserving the paints applied to Nanzuo’s POFP pottery.

Third, The POFP pottery from Nanzuo represents one of the earliest evidence of binding medium usage. Although the precise composition of this binding medium has yet to be confirmed, its presence at the Nanzuo, Dadiwan, and Taosi sites indicates that binding medium technology was employed during the Neolithic Age, potentially influencing subsequent painted cultural artifacts.

Last but not least, The majority of the POFP pottery featuring cinnabar as red paint, excavated in Nanzuo, is believed to constitute components of ceramic drums. This suggests that POFP potteries may have served a more diverse array of functions, thereby enhancing our understanding of the contextual usage of POFP pottery during the Neolithic Period in China.