An insight into the origin of elemental chromium in the lacquer of Qin terracotta warriors

The origin of elemental chromium for the archaeological weapons from the pits of Qin terracotta warriors in China has been highly controversial. Although previous studies have highlighted that the chromium on the surface of weapon originated from the contamination of surrounding lacquer, the exact origin of chromium in the lac-quer remains unclear. In this work, the measurement by inductively coupled plasma-mass spectrometer (ICP-MS) firstly confirmed that the elemental chromium was indeed contained in the archaeological Qin original lacquer. Nevertheless, the amount of elemental chromium in the Qin lacquer was as low as 0.0759 μg/mg, disclosing that it was impossible to artificially add extra refined chromium-containing substance to the lacquer in the preparation of the terracotta warriors. The soil from the archaeological site of Qin lacquer was found to have a chromium amount of 0.0660 μg/mg by ICP-MS. After the hygrothermal and soil-buried aging cycles for the lab-prepared lacquer, the surface and depth elemental analyses by time of flight-secondary ion mass spectrometer (TOF–SIMS) showed a gradient distribution of elemental chromium from the surface to interior of aged lacquer, indicating the migration and enrichment behavior of elemental chromium from the burial soil towards the lacquer. To explore the migration mechanism of elemental chromium, fluorescence imaging technique was employed in combination with Fourier transform infra-red spectrometry (FT-IR) and X-ray photoelectron spectroscopy (XPS) characterizations. The results revealed that cat-echol-containing fragments were formed during hygrothermal and soil-buried aging of lacquer and consequently coordinated with chromium ions, inducing the migration of elemental chromium towards the lacquer.


Introduction
The terracotta warriors pit within the mausoleum of the first Qin emperor (Qin Shihuang, 259-210 BC) is one of the greatest archaeological discoveries in the twentieth century and locates at Lintong District, Xi'an City, Shaanxi Province, China [1,2].At the beginning of the archaeological excavation of terracotta warriors pit in the 1970s, a large number of terracotta warriors and bronzes were unearthed [3][4][5].Intriguingly, the excavated bronze weapons, such as sword, dagger-axe, spears, halberds, battle-axe, scimitar and arrowheads, featured with delicate craftsmanship and sharp blade even after more than 2,000 years of burial [6].For a long time, the surface of bronze swords (Fig. 1) was recognized to undergo a treatment of chromate, providing the protection against rust and corrosion.It highlights the advancement of China's ancient metallurgical technology and the ingenuity of ancient Chinese people [7,8].
However, 27 out of 464 excavated bronze weapons were detected by X-ray fluorescence spectrometer (XRF) to contain elemental chromium with relatively low concentration of about 0.1%wt on each surface [7].For all bronze weapons, elemental tin with anti-corrosion property was identified in a high concentration ranging from 5%wt to 25%wt [8][9][10].Therefore, the recognition on the utilization of chromate coating to prevent bronze weapons from corroding is controversial.It was found that bronze weapons with their non-metallic parts (e.g., wooden hilts, scabbards) coated with lacquer were most likely to show elemental chromium on their surfaces [7,11].Based on the discovery of elemental chromium in the Qin original lacquer, a possibility that the elemental chromium on some surfaces of bronze weapons was derived from the contamination by lacquer was proposed.Nonetheless, further research is demanded as for the origin of elemental chromium in the lacquer, which is beneficial to the clarification about the confused ancient craftsmanship of bronze weapons and the lacquer, as well as the good preservation and restoration of bronze weapons.
Currently, there are two speculations regarding the origin of elemental chromium in the Qin original lacquer.The first speculation is that the elemental chromium was artificially added into the lacquer sap as an additive to accelerate the curing of lacquer [7].Although the elemental chromium was identified on the surface of Qin original lacquer, the extremely low level of elemental chromium and the lack of knowledge about the distribution characteristics of elemental chromium within the lacquer make this speculation implausible.The second hypothesis is that the presence of surface chromium of Qin original lacquer could be attributed to the contamination from the surrounding soil [12], because the lacquer employed as a decorative coating on the surfaces of non-metallic parts of the terracotta warriors and bronze weapons (Fig. 1) has been buried in the soil for over 2,000 years.However, there is little concrete evidence on the precise amount of elemental chromium in the soil itself and the experimental verification of chromium migration towards lacquer, as well as the elucidation of potential migration mechanism.
Herein, our research concentrated exclusively on responding these speculations about the origin of elemental chromium for the Qin original lacquer through qualitative-quantitative data, experimental evidences and mechanism analyses.As is known, non-destructive inspection of X-ray computed tomography (CT) allows material imaging to evolve from two dimensions to three dimensions [13][14][15], which facilitates the insight into the hierarchical structure of the Qin original lacquer.Non-destructive X-ray photoelectron spectroscopy (XPS) technique could distinguish imperceptible chemical groups in addition to the elemental composition on the top 10 nm of a film [16,17], which is advantageous to understand the surface chemical structure of the lacquer.Inductively coupled plasma-mass spectrometer Fig. 1 The burial and excavation status of bronze swords in the terracotta warriors pit (ICP-MS) method offers extremely low detection limits (parts per trillion concentration) for a wide range of minor and trace metal elements in solution [18,19], which is crucial to the accurate quantification of metal element concentration within the lacquer and soil in spite of a damage to the sample for ICP-MS analysis.Less destructive time of flight-secondary ion mass spectrometer (TOF-SIMS) technique is specialized in two-dimensional and three-dimensional chemical mapping of the surface distribution and in-depth distribution of chemical species [20][21][22], which enables the migration patterns of elemental chromium towards lacquer to be probed.Non-destructive fluorescent imaging visualization technique is highly sensitive for the early-stage identification of polymer aging pathways by using fluorescent labeling to specifically target aging-induced chemical groups [23,24], which makes it promising in monitoring ultra-traceable functional groups generated by anti-aging lacquer in the aging process.Based on the relation between the characteristic detection capabilities of mentioned analytical methods and the great challenges of our research, CT, XPS, ICP-MS, TOF-SIMS and fluorescent imaging techniques were newly introduced to investigate the origin of elemental chromium for the Qin original lacquer in combination with additional Fourier transform infrared spectroscopy (FT-IR) and scanning electron microscopy/ energy-dispersive X-ray spectroscopy (SEM-EDX).
The results strongly supported the viewpoint that elemental chromium in the Qin original lacquer was not artificially added.Moreover, catechol-containing fragments that could strongly interact with chromium ions were generated during the hygrothermal and soil-buried aging of lacquer, resulting in the migration of elemental chromium from the soil to the lacquer.In the future, the migration and enrichment behaviors of other metal elements towards the lacquer should also be emphasized, which contributes to optimizing soil management and implementing preventive protection and restoration for the relics from the pits of Qin terracotta warriors.

Materials
The Qin original lacquer and soil samples were from the corridor 10, test area 24, pit 1 of the terracotta warriors and provided by Emperor Qinshihuang's Mausoleum Site Museum.Specifically, the analyzed Qin original lacquer layer was not painted and directly in contact with the soil during the burial as shown in Fig. 1.The Qin original lacquer samples and the soil samples were collected immediately after excavation from the corridor 10.In addition to the terracotta warriors, dozens of bronzes were unearthed around the experimental samples.The lacquer saps for the preparation of lacquer samples in laboratory were from the lacquer tree in Ankang, Shaanxi Province, China.The NaCl (99.5 wt%), Na 2 SO 4 (99.0 wt%), HNO 3 (65% ~ 68%), HCl (36% ~ 38%), 3-(10-phenyl-9-anthracenyl) phenyl boronic acid (DPBA), Na 2 CO 3 and NaHCO 3 were obtained commercially and used as received without further purification.

Sample preparation
The Qin original lacquer was cleaned with brush, ultrapure water with a resistivity greater than 18 MΩ and dried in a vacuum oven at 30 °C for 24 h in sequence.The soil samples were first sieved to remove coarse particles and then ground to refine them.Finally, the samples were filtered through a 100-mesh sieve to ensure uniformity and fineness.To prepare the lacquer sample, the lacquer sap was coated onto a clean glass plate, followed by drying at 21 °C and 75%-80% R.H.The obtained lacquer film was denoted as laboratory-prepared (lab-prepared) lacquer.

Accelerated aging testing
The accelerated aging test was conducted on the lab-prepared lacquer in the constant temperature and humidity test chamber (TH-100A, Xi'an Qingsheng Electronic Technology Co., Ltd.).The lab-prepared lacquer samples were buried in the soil and then placed on a glass-made tray to allow for aging.Alternatively, the lab-prepared lacquer samples were wrapped by gauze saturated with the mixture comprising 0.5 mol/L NaCl and 0.5 mol/L Na 2 SO 4 solution and subsequently placed on an alloymade tray for aging.To achieve each aging cycle, the prepared samples were maintained at 35 °C and 50% R.H for 2 h, at 60 °C and 20% R.H for 4 h, and 50 °C and 95% R.H for 2 h in sequence.

Characterization
X-ray computed tomography (CT) of Carl Zeiss Xradia 610 Versa was utilized as a rapid, non-destructive and high-resolution three-dimensional tomographic analysis technique for Qin original lacquer.The main scanning parameters were listed as follows: a tube voltage of 50 kV, a tube current of 200 mA, an image resolution of 10 μm, an exposure time of 40 ms, a 360° scanning range with a rotation step size of 0.2°, and a total scanning time of 40 min.
Fourier-transform infrared (FT-IR) attenuated totalreflectance (ECO-ATR) spectra of the Qin original lacquer and lab-prepared lacquer samples were recorded on a Bruker VERTEX70 spectrometer.Specifically, the spectral range of sample was set from 400 cm −1 to 4000 cm −1 with a resolution of 4 cm −1 and the FT-IR ATR spectrum was available after 16 scans.
The surface elemental compositions of the Qin original lacquer and lab-prepared lacquer samples were investigated by Thermo Fisher ESCALAB Xi + X-ray photoelectron spectroscopy (XPS) using an Al mono Kα X-ray source.The XPS spectra was acquired at a spot size of 500 µm, an energy step size of 1.000 eV and the number of energy steps of 1361.Specifically, all binding energies were referenced to the C1s neutral carbon peak at 284.8 eV.
Nexion 350D inductively coupled plasma-mass spectrometer (ICP-MS) was employed to determine the metal element compositions and their contents in the Qin original lacquer, lab-prepared unaged lacquer and soil samples.Prior to ICP-MS test, the samples should be digested.In a typical run, the glass containers were cleaned by immersion in a 20% HNO 3 solution for 24 h.In a 25-mL of clean beaker, 0.0500 g of lacquer sample was boiled in a 1.0 mL of 18% HCl solution for 10 min.During the boiling of the lacquer sample, the ultrapure water with a resistivity greater than 18MΩ was added in moderation to ensure that the lacquer sample remained submerged in the solution.Afterwards, the resultant suspension was filtered.The collected remaining liquid was transferred into a 5-mL volumetric flask and brought to volume by ultrapure water.For the digestion of soil, the sample was first dried in an oven at 100 ℃ for 2 h, followed by grinding in an agate mortar and filtered through a 200-mesh sieve in sequence.For the digestion of soil, a 0.1000 g soil sample was treated with 5 mL of HF at 110 ℃ for 4 h, 12 mL of mixture comprising 8 mL of HNO 3 and 4 mL of HCl at 70 ℃ for 30 min in sequence.Afterwards, the acid-treated soil sample was further subjected to a high-pressure digestion in a microwave at 195 ℃ for 30 min.Finally, the obtained solution was concentrated to 1.0 mL for the ICP-MS analysis.Based on the sample pre-treatment procedures, the determination limits of ICP-MS method for detected elements in lacquer sample and soil sample were 1 ppb and 0.1 ppb, respectively.The final reported elemental contents were the averages of three different samples plus an error bar.
Time of flight-secondary ion mass spectrometer (TOF-SIMS), a M6 Plus device (ION-TOF GmbH, Germany), was introduced to acquire a three-dimensional distribution map of elemental chromium within the lacquer sample.To exclude the effect of contamination, the lacquer sample was treated with an O 2 sputter beam at 1 keV energy before starting the test.Afterwards, TOF-SIMS positive ion maps were collected over an area of 100 μm × 100 μm on the sample surface by using the primary ion source of Bi + pulse beam with the energy of 30 keV and a pulsed target current of 1 pA.
Surface morphologies and elemental compositions of the Qin original lacquer, lab-prepared unaged lacquer and aged lacquer samples were characterized by a scanning electron microscopy/energy-dispersive X-ray spectroscopy (SEM-EDX, Gemini500) at an acceleration voltage of 10 kV and 15 kV, respectively.Prior to the characterization, the surfaces of lacquer samples were coated with gold.
The fluorescence imaging of lacquer was performed on a Leica DMi8 fluorescence microscope equipped with four fluorescence channels.The DPBA was used for the fluorescent labeling towards hydroxyl groups generated on the surface of lacquer.Typically, the dye solution of DPBA (0.10 mmol/L) was prepared by dissolving 3.7 mg of DPBA in 30.0 mL of methanol and then diluting with 70.0 mL of Na 2 CO 3 -NaHCO 3 buffer solution (pH = 10.5).Subsequently, the lacquer samples after different aging durations were separately immersed in a 0.10 mmol/L DPBA solution and subjected to ultrasonic treatment for 2 min.Finally, the samples were sonicated in methanol for additional 10 min to remove any free DPBA molecules, followed by air-drying for fluorescence imaging at specific DAPI fluorescence channel.

Results and discussion
The surface microstructure analysis of lacquer The possible migration of elemental chromium in the pits of terracotta warriors involves the surface and interfacial behaviors.Therefore, the surface morphologies and chemical groups of Qin original lacquer and labprepared unaged lacquer samples were firstly characterized and compared.It is well-known that urushiol in the lacquer sap is oxidized by laccase in the presence of O 2 to produce a biphenyl dimer and then polymerized via side-chain C = C bonds of urushiol.Both oxidation and polymerization reactions proceed continuously, resulting in the formation of lacquer film with complex crosslinked structure and abundant C-C and C-O-C bonds (Fig. 2).SEM observation in Fig. 3a reflects that the surface of lab-prepared unaged lacquer is smooth but comprises a few circular holes with uneven sizes in the range 0.1-0.6 μm originating from the presence of air bubbles within the lacquer sap during the film formation process.For frail Qin original lacquer sample investigated in our work, the cross-sectional CT image in Fig. 3b first discloses that it possesses a two-layer structure with thickness of each lacquer layer ranging from the 25 μm to 40 μm, hinting a double-layer lacquering process in the preparation of the terracotta warriors.In addition, CT image reveals that both layers of lacquer films are unrobust.Specifically, the surface of lacquer layer with more fragile structure was elaborately observed by SEM and is found to be uneven and powdery in Fig. 3c, indicating that the Qin lacquer is severely degraded after more than 2,000 years of burial.
Owing to the formation mechanism of lacquer in Fig. 2, the FT-IR ATR spectrum of lab-prepared unaged lacquer sample in Fig. 4 shows the typical peaks at 2922 cm −1 and 2853 cm −1 assigned to stretching vibration of C-H bonds from -CH 3 and CH 2 groups, as well as the peaks at 1606 cm −1 and 1452 cm −1 assigned to stretching vibration of C=C-C bonds from phenyl rings.In addition, a weak peak at 3397 cm −1 is observed, which should be attributed to stretching vibration of O-H bonds from residual urushiol.For the FT-IR ATR spectrum of Qin original lacquer in Fig. 4, the peaks due to the stretching vibration of C-H bonds are also presented and locate at the same position as that in FT-IR ATR spectrum of labprepared unaged lacquer.Differently, the observed peak of O-H bonds is obviously strong and broad, implying that a large number of -OH groups are generated for the Qin lacquer sample in the burial environment.As a result of the O-H•••O hydrogen bond interaction between -OH groups on the phenyl rings, the peaks assigned to stretching vibration of C=C-C bonds from phenyl rings shift towards low wavenumbers at 1567 cm −1 and 1404 cm −1 , respectively.
XPS analysis is further used to provide insight into the surface (1-3 nm) chemical composition of Qin original lacquer and lab-prepared unaged lacquer samples.As shown in Fig. 5a, the preliminary wide scanning spectra show that the surfaces of both samples mainly consist of carbon and oxygen elements.In addition, traces of nitrogen, calcium and lead elements were also detected on the surface of Qin lacquer sample.To identify the dominant chemical states of carbon on the surfaces of both lacquer samples, high-resolution XPS spectra of C1s peaks were obtained after the subtraction of Shirley-type background.In Fig. 5b, the high-resolution C1s spectrum of lab-prepared unaged lacquer sample contains five different peaks at 284.8 eV, 286.5 eV, 287.9 eV, 288.

The elemental compositions of metals in the lacquer and soil
The metal element compositions and their contents in the Qin original lacquer and lab-prepared unaged lacquer samples were determined by ICP-MS after the semidigestion of samples.Metal elements with high content besides chromium in the lacquer samples are listed in Fig. 6a.It shows that both lacquer samples mainly contain elemental sodium, magnesium, aluminum, potassium, calcium, iron and copper.In contrast, the contents of these elements in the Qin original lacquer are all significantly higher than those in the lab-prepared unaged lacquer samples.In regard to the elemental chromium in Fig. 6b, its content in the Qin original lacquer sample is 0.0759 μg/mg, whereas the content in the lab-prepared unaged lacquer sample is 0.0001 μg/mg.It should be emphasized that the actual chromium contents in both lacquer samples are higher than those detected by ICP-MS, because both lacquer samples were not thoroughly digested.Nevertheless, the measured trace amounts of elemental chromium in both lacquer samples by ICP-MS could confirm that it is impossible to artificially add extra chromium-containing substance to the lacquer in the preparation of the terracotta warriors.
Yet it is worth noting from the ICP-MS result that the chromium content in the Qin original lacquer sample is 516 times higher than that in the lab-prepared unaged lacquer sample.One possible speculation about the high chromium content in the Qin lacquer sample is that the elemental chromium leaches easily from the Qin lacquer sample during the digestion, because the internal crosslinked structure of Qin lacquer sample was destroyed in the burial environment.Without the digestion, TOF-SIMS technique is proposed as an alternative to determine the chromium content.However, the secondary ion map provided by TOF-SIMS in Fig. 7a shows almost no counts for the Cr + ions across the entire surface of lab-prepared unaged lacquer.Moreover, the depth distribution curve of Cr + in Figs.7c and three-dimensional (3D) image of Cr + in Fig. 7d also highlight that no Cr is detected in the lab-prepared unaged lacquer being its signal below the detection limit of TOF-SIMS.Conversely, TOF-SIMS secondary ion map in Fig. 7b shows a uniform and dense distribution of Cr + counts across the entire surface of Qin original lacquer.In addition, the elemental chromium is relatively evenly distributed from the surface of Qin original lacquer to the interior, as revealed by the depth distribution curve of Cr + in Figs.7c and 3D image of Cr + in Fig. 7e.Importantly, CrO + is also found in the Qin original lacquer and its depth distribution curve is displayed in Fig. 7c.Despite a low signal intensity, CrO + shows an even distribution from the surface to the interior like Cr + .Therefore, TOF-SIMS result further confirms that the Qin original lacquer itself indeed has higher chromium content than the lab-prepared unaged lacquer.Moreover, Cr-O bonds are present in the Qin original lacquer.This phenomenon implies that the migration and enrichment of elemental chromium occurred for the lacquer in the burial environment.As shown in Fig. 6b, the soil from the archaeological site of Qin lacquer has a chromium amount of 0.0660 μg/mg.Attributed to the non-contact with soil after excavation, one possible speculation is that the elemental chromium in the burial soil gradually migrates towards the inside of lacquer before excavation.

The migration behavior of elemental chromium
To investigate the migration behavior of elemental chromium from the soil towards lacquer, the accelerated hygrothermal aging experiment was performed onto the lab-prepared lacquer buried in the soil.Firstly, SEM is used to characterize the surface morphologies of aged lacquers after multiple aging cycles.As can be seen in Fig. 8, the edges of circular holes on the surface of lacquer gradually peel off and get powder with increasing number of aging cycles, implying that the deterioration of lacquer during the accelerated aging experiment develops from the defects of circular holes on the surface.In addition, EDX elemental composition analysis attached to the SEM shows that the atomic percentage of elemental chromium on the surface of unaged lacquer, the lacquer after 30 aging cycles and the lacquer after 60 aging cycles is 0.70%, 1.05% and 1.34%, respectively.The enhancement of chromium content with the increase in the number of aging cycles seems to verify the migration behavior of elemental chromium from the soil towards lacquer.However, the content of elemental chromium is not detected on the surface of lacquer after 90 aging cycles, which should be attributed to the extremely reduced sensitivity of EDX on the rough surface shown in Fig. 8d.Therefore, the TOF-SIMS technique, rather than EDX, is employed again to investigate the migration behavior of elemental chromium.
Figure 9 shows the TOF-SIMS secondary ion maps of Cr + counts from the lab-prepared lacquer after 30, 60 and 90 hygrothermal and soil-buried aging cycles, respectively.It is found that strong Cr + counts present on each surface of aged lacquer and exhibit a micro-regional distribution pattern, which increases with increasing number of aging cycles.This indicates that elemental chromium is continuously enriched from the soil to the surface of lacquer during aging.Intriguingly, the emergence of micro-regional distribution pattern of Cr + counts is similar to the appearance of circular micronsized hole-like morphology on the surface of aged lacquer in Fig. 8, implying that the surface enrichment of elemental chromium is mainly concentrated at the developed hole-like defects during the aging of lacquer.
In addition, the corresponding TOF-SIMS depth distribution curves of Cr + for three lacquer samples after hygrothermal and soil-buried aging are depicted in Fig. 10a.Obviously, the intensity of Cr + signal is gradually attenuated from the surface to the inside for each sample, implying a gradual inside diffusion and penetration of surface enriched elemental chromium.On the other Fig. 6 a ICP-MS results for the metal element compositions and their content in both lacquer samples; b ICP-MS results for the chromium content in the lacquer and burial soil sample hand, it is found that the intensity of Cr + signal within the aged lacquer enhances as the number of hygrothermal and soil-buried aging cycle increases.As highlighted in Fig. 10b-d, TOF-SIMS 3D images visually illustrate the concentration gradient distribution of Cr + from the surface to the inside of aged lacquer and the enhancement in Cr + concentration with increasing cycle number.In addition, it should be also pointed out that CrO + with lower signal intensity than Cr + is found in the lab-prepared lacquers after aging cycles.Especially, TOF-SIMS 3D image of CrO + after 90 aging cycles is displayed in Fig. 10e.In brief, TOF-SIMS results confirm that the lacquer could effectively adsorb and accumulate elemental chromium from surrounding burial soil.Prolonging soil-buried durations for lacquer could strengthen the surface enrichment of elemental chromium and even internal penetration.This further confirms the migration of elemental chromium during the soil burial of Qin original lacquer before excavation.Owing to the behavior described above, the Qin original lacquer subjected to more than 2,000 years of burial ultimately yields a more uniform distribution of elemental chromium on the surface and within its interior.
To further investigate the adsorption capacity of elemental chromium of the lab-prepared lacquer, additional accelerated hygrothermal aging experiment was conducted on the lacquer in contact with alloy using NaCl and Na 2 SO 4 as a medium.The surface of alloy contains more elemental chromium of 11.88 wt%, which could be easily leached out in the presence of NaCl and Na 2 SO 4 under hygrothermal condition.Specifically, the lab-prepared lacquers were also subjected to hygrothermal and alloy-contacted aging for 30, 60 and 90 cycles, respectively.Afterwards, obtained depth distribution curves of Cr + and 3D images of Cr + by TOF-SIMS for these three aged lacquer samples are illustrated in Fig. 11.Clearly, the trend of surface enrichment and internal penetration of elemental chromium for lacquer  during the hygrothermal and alloy-contacted aging is the same as that during the hygrothermal and soil-buried aging.Nevertheless, the comparison between Figs. 10 and Fig. 11 demonstrates that the ability of the lacquer to enrich elemental chromium from chromium-rich alloy is greater than that from chromium-poor soil under hygrothermal condition.Therefore, it can be concluded that the lacquer could adsorb and enrich elemental chromium from the chromium-containing source nearby during the hygrothermal aging process.In addition, the ability of lacquer to enrich elemental chromium is dependent on the chromium content of the source nearby.

The migration mechanism of elemental chromium
It is of great significance to disclose the migration mechanism of elemental chromium towards the lacquer in the soil-buried environment, so as to scientifically explain the origin of elemental chromium for the archaeological relics, as well as to predict and control the potential enrichment of elemental chromium in the protective treatment of relics from the pits of Qin terracotta warriors.To achieve migration mechanism of elemental chromium, an in-depth understanding of surface groups for the lacquer during the hygrothermal and soil-buried aging is crucial.In this work, the generation of surface groups during aging was detected and visualized by the fluorescent probe that could target a certain functional group specifically and sensitively.Aging for polymers is manifested as the generation of different functional groups, such as -OH, -COOH and -NH 2 groups.As revealed by FT-IR ATR spectrum in Fig. 4, a large number of -OH groups are generated for the Qin lacquer sample after more than 2,000 years of burial.Therefore, the generation of -OH groups during the hygrothermal and soil-buried aging was speculated for lacquer and was identified by specific DPBA fluorescent (-OH)-targeting molecule.In Fig. 12ac, two-dimensional fluorescence confocal microscopy images are obtained from the lab-prepared lacquers after hygrothermal and soil-buried aging for 30, 90, and 150 cycles, respectively.As expected, blue-emissive blocks are seen for each aged lacquer sample, representing the existence of -OH groups.Particularly, the quantities of blue-emissive blocks are larger for the surface of lacquer after more aging cycles.In contrast, blue-emissive blocks could not be seen for the unaged lab-prepared lacquer labeled with DPBA.More elaborately in this work, specific fluorescent (-COOH)-targeting molecule was also employed to label potential generation of carboxyl groups from these aged lab-prepared lacquers but corresponding emissive blocks was almost not discerned.These results verify that the -OH groups are generated during the hygrothermal and soil-buried aging for the lacquer.Moreover, the regional distribution of blue-emissive blocks indicates that the generation of -OH groups is inclined to occur at the regional physical hole-like defects shown in Fig. 8.
Catechol-containing groups are known to chelate transition metal ions (e.g., Cr 2+ , Cu 2+ and Fe 3+ ) and other metal ions (e.g., Na + , Mg 2+ , Ca 2+ and Al 3+ ) by noncovalent interactions or chemical bonds.Attributed to the strong interactions of metal-catechol and the identified -OH groups on the surface of aged lacquer, possible migration and enrichment mechanism for the elemental chromium towards the lacquer is proposed in Fig. 12d.The phenoxy groups in lacquer (shown in Fig. 2) are apt to break to form -OH groups under hygrothermal and soil-buried aging, yielding catechol-containing fragments and consequently triggering efficient interactions with chromium ions.In this case, the migration and enrichment of elemental chromium towards the lacquer is induced to form catechol-chromium complexes.In fact, previous studies have confirmed that the Ca 2+ has great coordination ability for catechol-containing groups [25].Coincidentally, the experimental fact in Fig. 6a that the content of elemental calcium is the highest in the lacquer also proves the preferred interaction between elemental calcium and catechol-containing groups according to our proposed migration mechanism.In addition, Fig. 6a shows that the elemental Na, Mg, Al, Fe and Cu in the Qin original lacquer sample are all higher than those in the lab-prepared unaged lacquer sample in the case of chelation-induced migration and enrichment.Therefore, the proposed mechanism for the migration of elemental chromium towards the lacquer in this work has demonstrated its certain applicability in explaining experimental findings and predicting the possible migration behavior of metallic element.

Conclusions
To definitively address the origin of elemental chromium in the archaeological Qin original lacquer from the pits of Qin terracotta warriors in China, a series of analytical strategies was employed to provide a detailed and conclusive evidence.
Preliminarily, the CT and SEM measurements showed a bilayer structure with uneven and powdery surface for Qin original lacquer.The FT-IR and XPS techniques revealed that the surface of Qin original lacquer contained more -OH groups but less C-O groups in comparison with the surface of lab-prepared unaged lacquer.After detailed characterization of the metal elements in Qin lacquer by ICP-MS with low detection limit, the presence of elemental chromium with a content of 0.0759 μg/mg was confirmed.More importantly, the surface and depth elemental analyses by TOF-SIMS manifested that the Qin original lacquer had higher chromium content than lab-prepared unaged lacquer and the elemental chromium in Qin original lacquer showed a gradient distribution from the surface to interior.The findings of the ultra-low amount of elemental chromium in the Qin original lacquer sample strongly supported the viewpoint that elemental chromium was not artificially added into the Qin lacquer.
To explain the higher chromium content in Qin original lacquer, the accelerated hygrothermal and soil-buried aging experiments were conducted on the lab-prepared lacquers.The result from the TOF-SIMS characterization indicated that elemental chromium in the soil continuously migrated and enriched on the surface and inside of the lab-prepared lacquer as the aging time increased.Moreover, it was found that the elemental chromium was preferentially enriched at the physical defects of aged lacquer and the ability of lacquer to enrich elemental chromium relied on the chromium content of the chromium-containing source nearby.Significantly, the possible migration mechanism for the elemental chromium towards the lacquer was investigated by labeling the surface groups from the lacquer aging process through specific fluorescent probes.The results disclosed the generation of -OH groups at the regional physical defects during the hygrothermal and soil-buried aging for the lacquer.Combined with the chemical structure of lacquer, catechol-containing fragments were speculated to form.Attributed to the strong interaction between the catechol-containing groups and metal ions, the migration and enrichment of elemental chromium towards the lacquer were induced.Notably, the explanation for the highest calcium content in Qin original lacquer using our proposed element migration mechanism was consistent with the reported results, indicating the practical applicability of newly proposed mechanism.
6 eV and 289.5 eV, corresponding to five different types of carbon in C-C/C=C, C-O, C-O-C, C=O and COO groups, respectively.In contrast, the high-resolution C1s spectrum of Qin lacquer sample in Fig. 5c includes three distinct peaks at 284.8 eV, 286.7 eV and 288.6 eV, corresponding to three different types of carbon in C-C/ C=C, C-O and C=O groups, respectively.In addition, it is worth noting that the results from high-resolution C1s

Fig. 2
Fig. 2 Schematic illustration representing the formation of lacquer film

Fig. 3 a
Fig. 3 a SEM image from the surface of lab-prepared unaged lacquer; b-c CT and SEM images from the cross-section and surface of Qin original lacquer

Fig. 5 a
Fig. 5 a XPS survey spectra of lab-prepared unaged lacquer and Qin original lacquer samples; b, c high-resolution C1s spectra of corresponding lacquer samples

Fig. 7
Fig. 7 TOF-SIMS secondary ion maps of Cr + counts from the lab-prepared unaged lacquer (a) and the Qin original lacquer (b); depth distribution curves of Cr + and CrO + (c); 3D images of Cr + (d and e)

Fig. 8
Fig. 8 Surface morphologies of a the lab-prepared unaged lacquer; b the lacquers by hygrothermal and soil-buried aging for 30 cycles, c for 60 cycles and d for 90 cycles

Fig. 9
Fig. 9 TOF-SIMS secondary ion maps of Cr + counts from the lab-prepared lacquers by hygrothermal and soil-buried aging for 30 cycles (a), 60 cycles (b) and 90 cycles (c)

Fig. 10 aFig. 11 a
Fig. 10 a TOF-SIMS depth distribution curves of Cr + and b, c and d 3D images of Cr + from the lab-prepared lacquers after hygrothermal and soil-buried aging for 30 cycles, 60 cycles and 90 cycles in sequence; e 3D image of CrO + after 90 aging cycles