Characterization of the rubbing sample by using SEM-EDS
The morphology characteristics of ink on the historical rubbing sample were observed by scanning electron microscopy (SEM) at a magnification of 20000 times. The results are shown in Fig. 2. As can be seen, the ink particles were closed to circle. The ink particles were not uniform in size, the large particles were about hundreds of nanometers, and the small particles were dozens of nanometers. This feature is consistent with the characteristics of pine wood soot ink, which give a clue that the ink used for the rubbings is probably pine wood soot ink [15]. Due to the limit of this method, further investigation by Py-GC/MS was carried on, which is stated in the following subsection.
The mapping analysis of scanning electron microscope and energy dispersive spectrometer (SEM-EDS) was used to analyze the particulate in the paper in the historical rubbing sample. The results are shown in Fig. 3. It can be clearly seen that the particles primarily contain Al, Si and Ca elements, which confirmed that kaolin (a kind of clay, molecular: Al203·2Si02·2H20) is present in the rubbing sample, which is used as fillers in papermaking process to improve the whiteness, smoothness and opacity of paper according to the literature [2].
Characterization of the historical rubbings by using Py-GC/MS
Py-GC/MS was utilized to analyze the historical rubbing sample, the chromatogram obtained is presented in Fig. 4, while the primarily compounds identified are listed in Table 1. Five types of substances were detected in the historical rubbing sample by using Py-GC/MS analysis. (1) a series of polycyclic aromatic hydrocarbons (PAHs) could be identified as the follow compounds: anthracene, fluoranthene, pyrene, triphenylene, benzo[k]fluoranthene and benzo[ghi]perylene, labeled as S1, S2, S3, S4, S5, and S6 in Fig. 4 respectively. The PAHs were from soot, in which the relative high content of S5 and S6 could be used as an indication of the presence of certain soot ink [16]; (2) a small number of compounds including retene (as show in Fig. 4, labeled as T1) and methyl dehydroabietate (as shown in Fig. 4, labeled as T2) were detected, which were from conifer tar according to the literature [17], the chemical structure of the two compounds were shown in figure 5. (3) In Fig. 4, G1, G2 and G3 represent 3-methyl-1H-pyrrole, indole, and 3-methyl-indole respectively, which were the pyrolysis marker compounds of egg white [[i], [ii]]. In addition, 3-ethoxy-, (3beta)-Cholest-5-ene (labeled as G4) was also detected, which is the main component of egg yolk; (4) Compounds of menthol and curcumene were detected in the rubbings (labeled as E1 and E2 in the chromatogram). Menthol is the main chemical component of peppermint oil, which is obtained from the roots, stems and leaves of peppermint by steam distillation. It is an important flavor additive [[iii]]. Curcumene is the main chemical component of the essential oil of turmeric herbs, which has antibacterial, antitumor and antioxidant effects [[iv], [v]]. (5) furthermore, furan structure (as shown in Fig. 4, labeled as 3), ketones (as shown in Fig. 4, labeled as 2 and 5), carbohydrate (as shown in Fig. 4, labeled as 10) and a series of compounds characterized by a base peak at 218 m/z, corresponding to the beta-amyrin (as shown in Fig. 4, labeled as 13) and alpha-amyrin (as shown in Fig. 4, labeled as 14) were noticed, which were the pyrolysis marker compounds of bark fiber [9], representing the paper probably made from mulberry bark (Morus alba L.). In order for the unambiguous identity of the fiber, Herzberg staining method will be carried out in following.
[i]. Giuseppe C, Guido CG, Lanterna G, et al. The potential of pyrolysis-gas chromatography/mass spectrometry in the recognition of ancient painting media. Journal of Analytical and Applied Pyrolysis. 1993;(24):227-242.
[ii]. Wang N, Gu A, Min JR, Li GH, Lei Y. Identification of Protein Binding Media Used in Chinese Cultural Relics by Pyrolysis-Gas Chromatography/Mass Spectrometry Chinese. J. Anal. Chem. 2020;48(1):90-96.
[iii]. Guo XK. Bo he you hua xue cheng fen fen xi [Analysis of chemical constituents of peppermint oil]. Special economic animal and plant. 2020;23(9):33-35.
[iv]. Yan H, Li PH, Zhou GS, et al. Quality evaluation of different origins and specifications of zingiberis rhizoma medicinal materials and decoction pieces. Journal of Chinese Medicinal Materials. 2020;43(4):817-823.
[v]. Qiang YY, Wei H, Fang L, et al. Analysis of chemical components of volatile oil in turmeric (Curcuma longa L.) from Fujian by HS-SPME-GC-MS. China Food Additives. 2020; 31(1):147-153.
Table 1. List of compounds of the rubbing paper with retention time (RT), expected main fragment ions, most likely attribution of the products, and relative content.
No
|
RT(min)
|
Main Ions
|
Identified compounds
|
Area (%)
|
G1
|
3.35
|
53, (80)
|
1H-Pyrrole, 3-methyl-
|
0.84
|
1
|
5.32
|
51, 78, (104)
|
Styrene
|
1.15
|
2
|
5.85
|
53, (67), 96
|
2-Cyclopenten-1-one, 2-methyl-
|
1.6
|
3
|
8.57
|
53, 67, (96)
|
2,4-Dimethylfuran
|
2.49
|
4
|
9.51
|
66, (94)
|
Phenol
|
1.8
|
5
|
11.61
|
(67), 95, 110
|
2-Cyclopenten-1-one, 2,3-dimethyl-
|
1.26
|
E1
|
12.61
|
69, (109), 119, 134
|
cis-p-mentha-1(7),8-dien-2-ol
|
0.82
|
6
|
13.35
|
79, (108)
|
3-methyl-Phenol
|
2.73
|
E2
|
16.98
|
71, (81), 95, 123, 138
|
Menthol
|
0.61
|
7
|
18.86
|
(91), 131
|
Benzenepropanenitrile
|
1.54
|
G2
|
20.57
|
63, 90, (117)
|
Indole
|
2.38
|
G3
|
23.29
|
77, 103, (130)
|
3-methyl-Indole
|
1.8
|
E3
|
25.58
|
105, (119), 132, 202
|
Curcumene
|
0.4
|
8
|
26.2
|
(55), 69, 83, 97, 111
|
1-Pentadecene
|
1.01
|
9
|
28.73
|
(55), 69, 83, 97, 111
|
1-Heptadecene
|
0.89
|
10
|
30.29
|
(60), 73
|
1,6-anhydro-bete-D-Glucopyranose
|
26.5
|
S1
|
33.2
|
76, 152, (178)
|
Anthracene
|
0.38
|
11
|
36.2
|
(74), 87, 143, 270
|
Hexadecanoic acid, methyl ester
|
4.09
|
S2
|
39.01
|
101, (202)
|
Fluoranthene
|
0.24
|
S3
|
40.05
|
101, (202)
|
Pyrene
|
0.26
|
12
|
40.18
|
(74), 199, 255,298
|
Methyl stearate
|
4.96
|
T1
|
41.9
|
189, 204, (219), 234
|
Retene
|
0.22
|
T2
|
44.02
|
(239), 299, 314
|
Methyl dehydroabietate
|
0.12
|
S4
|
46.07
|
101, 113, (228)
|
Triphenylene
|
0.15
|
S5
|
52.33
|
113, 125, (252)
|
Benzo[k]fluoranthene
|
0.19
|
G4
|
52.77
|
(57), 95, 147, 353, 368
|
3-ethoxy-, (3.beta.)-Cholest-5-ene
|
0.21
|
13
|
54.66
|
95, 135,189, 203,(218)
|
.beta.-Amyrin
|
0.18
|
14
|
55.3
|
95, 135,189, 203,(218)
|
.alpha.-Amyrin
|
1.06
|
S6
|
57.87
|
138, (276)
|
Benzo[ghi]perylene
|
0.29
|
In order to see the relative contents of the PAHs, analyses by Py-GC/MS with select ion mode (SIM) for S1-S6 were conducted. The chromatogram obtained is depicted in Fig. 6. For comparison, the relative content of the PAHs of the reference materials of the pine wood soot ink (PM), lamp soot ink (LM) and carbon black (CB), which were published earlier [16], as well as the PAHs results of the historical rubbings (RM) are listed in table 2. The relative content of benzo[k]fluoranthene and benzo[ghi]perylene in the historical rubbing sample is 31.78% (peak area), and the triphenylene is 9.93% (peak area), which is in accordance with that of pine wood soot ink. Combining the detection of marker compounds of retene and methyl dehydroabietate, pine wood soot ink used for the rubbings can be concluded.
Table 2. The relative contents of the main PAHs in the modern pine wood ink (PM), lamp soot ink (LM), carbon black (CB) and the ink used in the historical rubbings (RM) by Py-GC/MS analysis in SIM mode
Sample
PAHs
|
PM
|
LM
|
CB
|
RM
|
S1 area(%) (m/z 178)
|
13.9
|
14.5
|
71.4
|
25.2
|
S2 area(%) (m/z 202)
|
27
|
34.3
|
9.8
|
15.89
|
S3 area(%) (m/z 202)
|
28.9
|
34.7
|
9.1
|
17.22
|
S4 area(%) (m/z 228)
|
10
|
3.8
|
0.0
|
9.93
|
S5/S6 area(%) (m/z 252)
|
20.2
|
12.5
|
9.5
|
31.78
|
In summary, the ink used for the rubbings is pine wood soot ink, with egg (egg white and egg yolk) as binding media, while peppermint oil and turmeric herbs oil are probably used as additives in the ink. No surprise, the finding of pine wood soot ink in the rubbings, since its use can date back as early as Han Dynasty (202 BC- AD 220) [11], which was continuously in use in the following Dynasties. Only after 11th century, lamp soot ink was gradually in place of pine wood soot ink. Animal glue is the common binding media used in the Chinese ink [12], However, in this case whole egg were found as binding media. Interestingly, peppermint oil and turmeric herbs oil are found as additives in the ink.
Fiber identification using Herzberg staining method
The morphological characteristics of a modern handmade mulberry bark paper are shown in Fig. 7(a) after Herzberg staining method, which was used as a reference sample. The fibers taken from the historical rubbing sample were examined under the paper fiber analyzer after Herzberg staining method. The results are shown in Fig. 7 (b). Although the fiber morphology characteristics were not obvious due to the presence of ink, however similar characteristics as the reference mulberry bark could be observed, including yellow amorphous waxy substance attached to the fiber and cell cavity, as well as the reddish brown color of the fibers. In addition, most of the fiber width is between 10-15μm, and the fiber length was about 6mm. These characteristics were identical as the morphological characteristics of the mulberry bark paper fiber [[i]]. Combining with the Py-GC/MS results about the detection of bark marker compounds, it could be determined that historical rubbings paper is mulberry bark paper.
[i]. Wang, J.H. 1999. Papermaking Raw Materials of China an Atlas of Micrographs and the Characteristics of Fibers, Beijing: China Light Industry Press P 129.
According to the Book of the Later Han (Cai Lun’s biography), the raw materials for papermaking included ramie rags, fishing nets, and paper mulberry during the Eastern Han Dynasty. During the Tang Dynasty, kozo, mulberry bark, and rattan bark were the primary sources for papermaking. The use of bamboo to make paper was primarily developed in Song Dynasty [[i]]. The identification of mulberry bark as the origin source of the historical rubbings paper, could provide another positive evidence for its authenticity, since the rubbings was recorded made from Tang Dynasty.
[i]. Wang J, Li Y. Cong ji zhong han zhi de fen xi jian ding shi lun wo guo zao zhi shu de fa ming [a]. Cultural Relics. 1980; 26(1): 80-84.