Journal of Cultural Heritage 32 (2018) 22–29

Available online at

ScienceDirect
www.sciencedirect.com

Original article

New results in Dead Sea Scrolls non-destructive characterisation.

Evidence of different parchment manufacture in the fragments from
Reed collection
Marina Bicchieri a,∗ , Armida Sodo a,b , Ira Rabin c,d , Anka Kohl c , Giovanna Piantanida a,e
a
Laboratorio di Chimica, Istituto Centrale per il Restauro e la Conservazione del Patrimonio Archivistico e Librario, Via Milano 76, 00184 Roma, Italy
b
Dipartimento di Scienze Università degli Studi Roma Tre, Via della Vasca Navale 84, 00146 Roma, Italy
c
Bundesanstalt für Materialforschung und -prüfung, Unter den Eichen 44-46 BAM 4.5, 12203 Berlin, Germany
d
Centre for the Studies of Manuscript Cultures, Hamburg University, Warburgstr. 36, 20354 Hamburg, Germany
e
Banca d’Italia, Servizio Organizzazione, Via Otricoli 41, 00181 Roma, Italy

a r t i c l e i n f o a b s t r a c t

Article history: This work presents the non-destructive spectroscopic characterisation of original Dead Sea Scrolls (DSS)
Available online 9 February 2018 parchment fragments from Ronald Reed collection. The fragments are of paramount importance because
they have never been subjected to any treatment of preservation and restoration, this allows to investi-
Keywords: gate the manufacturing method of real original Jewish parchments. The manufacture of “sacred” Jewish
Dead Sea Scrolls parchments, in fact, is traditionally supposed to use a superficial tannin treatment. To study the DSS frag-
Tannins ments, it was necessary both to analyse mock-up samples, especially manufactured in order to reproduce
M-Raman
ancient Oriental Jewish ritual parchments, and to compare the results with those obtained in the analysis
FT-Raman
ATR-FTIR
of modern and ancient Western Jewish ritual parchments, in order to test the effectiveness of the selected
spectroscopic techniques. Traditionally, the main difference between Oriental and Western traditional
parchment preparation consisted in the dehairing method: enzymatic for Oriental and lime-based for
Western. Moreover, a finishing treatment with tannin was supposed to be applied on ritual Jewish parch-
ments. The need of reference samples derives from the knowledge that each parchment preparation,
treatment and degradation can induce structural modifications that affect the spectral features. Fourier
Transform Infrared Spectroscopy by Attenuated Total Reflection (ATR-FTIR), FT-Raman and m-Raman
were used in this study. The experimental results allowed us to recognise, with different sensitivity, the
presence of tannin by using m-Raman and IR spectroscopies and to prove that not all the archaeological
samples were manufactured in the same way with vegetal extracts. Many salts (tschermigite, dolomite,
calcite, gypsum and iron carbonate) were found on the surface of DSS fragments. They can derive from
the degradation processes and storage environment before the discovery or from the manufacture. More-
over, the different sensitivities and instrumental characteristics of the used techniques permitted us to
establish an analytical protocol, useful for further studies of similar materials.
© 2018 Elsevier Masson SAS. All rights reserved.

1. Introduction public. The larger part of the collection is written on a skin-based

Ancient Jewish writings known today as Dead Sea Scrolls were
discovered shortly after World War II in the caves of the Judean
desert along the western shore of the Dead Sea. Dated to the end
of the Second Temple period and the “time of the Christ” they have
never stopped attracting attention of the scholars and the broad
∗ Corresponding author. Tel.: +390648291217.
E-mail addresses: marina.bicchieri@beniculturali.it (M. Bicchieri),
sodo@fis.uniroma3.it (A. Sodo), rabin@fhi-berlin.mpg.de (I. Rabin),
anka.kohl@bam.de (A. Kohl), giovanna.piantanida@gmail.com (G. Piantanida).
material that was rather soon associated with the skins that were
specially prepared for writing and were mentioned in Talmud.
The scholars who worked on the collection were greatly inter-
ested in the precise dates of the compositions and they turned to
Ronald Reed from Leeds University, an acknowledged expert in all
matters concerning ancient leather and parchment. They hoped
that the material itself might provide clues to dating, other than
palaeography. In 1956, when the first contact was made, the idea of
radiocarbon dating was just being born and could not be considered
as a practical alternative. The archaeologists chose a representative
collection of fragments bare of writing for the studies at the Leeds

https://doi.org/10.1016/j.culher.2018.01.014
1296-2074/© 2018 Elsevier Masson SAS. All rights reserved.
 M. Bicchieri et al. / Journal of Cultural Heritage 32 (2018) 22–29
University. This first material study of the scrolls took several years
and resulted in a number of scientific publications [1–3] and the
very detailed doctoral thesis of Poole [4]. Using wet chemistry, light
and electron microscopy and UV spectroscopy, the team concluded
that the collection contained leather and parchment,whereby the
latter was treated superficially with vegetable tannins, quite sup-
porting the description available in Talmud. Years later, in his book
on the ancient leather and parchment Reed revised his conclusions
and reported that one variety of the parchment had never been
tanned [5].
Today, the collection of the fragments studied by Reed and Poole
is a part of the Reed Archive in the John Rylands University Library
[6]. It was now made available to the authors of this paper for re-
investigation using new techniques that have been developed since
the first studies. The extreme importance of this collection relies
on the fact that none of the fragments was subjected to series of
consequent conservation treatments that considerable changed the
original material.
2. Aim
The aim of this work was the non-destructive characterization
of original DSS parchments from Reed collection by using molec-
ular vibrational spectroscopies. Moreover, it seemed important to
verify whether tannins application was a necessary step in the man-
ufacture of “sacred” Jewish parchment, as traditionally supposed.
Since tannins were applied in much smaller amounts than those
required for leather tanning, their unequivocal recognition by non-
destructive FTIR and Raman spectroscopies would lead to establish
a scientific protocol for a non-destructive characterization of skin-
based materials, both ancient Oriental Jewish ritual parchments
and modern and antique Western Jewish parchments.
This work reports the results obtained in the analysis of some
samples belonging to a larger number DSS fragments that were
considered as representative of the whole investigated materials.
Before moving to the original DSS fragments, mock-up samples
especially manufactured in order to reproduce ancient Oriental
Jewish ritual parchments were analysed, as well as modern and
ancient Western Jewish parchments, to test the effectiveness of the
selected spectroscopic techniques.
3. Materials and methods
The most representative analyzed samples are briefly described
below.
Mock-ups for the manufacturing method:
• GB1: goat parchment superficially treated with a 3% w/V solution
of gum Arabic in water; 10 mL were applied on a 20 × 20 cm2
surface. The final amount of added gum Arabic was 0.75 mg/cm2 ;
• GB2: goat parchment superficially treated with a 3% w/V solu-
tion of gum Raddiana (Vachellia tortilis sp.) in water; 10 mL were
applied on a 20 × 20 cm2 surface. The final amount of added gum
Raddiana was 0.75 mg/cm2 ;
• GB3: goat parchment treated superficially with oak gall extract at
15% w/V concentration in water. The galls were washed, grated
and cooked in a small amount of distilled water for 24 hours. The
resulting liquid was filtered and set to the final volume. 5 mL were
applied on a 20 × 20 cm2 surface. The final amount of added galls
extract was 1.875 mg/cm2 ;
• GB4: goat parchment treated superficially before with the oak
gall extract and then with Gum Arabic, at the same amount and
concentration used for GB1 and GB3 samples.
Modern and ancient Western Jewish sheep parchments:
23
• S1: modern ritual Jewish parchment manufactured with calcite
and treated with oak gall extract on the surface at unknown con-
centration;
• S2: modern ritual Jewish parchment treated with alum and oak
gall extract on the surface at unknown concentration;
• S3: ancient original ritual Jewish fragment of unknown manufac-
turing technique, 19th century;
• S4: ancient original ritual Jewish fragment of unknown manufac-
turing technique, 18th century.
S1 and S2 samples were specifically manufactured for the
restoration of sacred parchment belonging to the Roman Jewish
Community, courtesy of Rav. Amedeo Spagnoletto. Therefore, they
can be regarded as examples of the Western preparation of sacred
Jewish parchment.
Reed archive (R = Reed collection):
• R-4Q1 (Uninscribed fragment from Qumran, Cave 4);
• R-4Q2 (Uninscribed fragment from Qumran, Cave 4);
• R-4Q11 (Uninscribed fragment from Qumran, Cave 4);
• R-M9 (Uninscribed fragment from Cave Murabba’at);
• R-M10 (Uninscribed fragment from Cave Murabba’at);
• R-Q9 (Uninscribed fragment from unknown cave);
• R-A45 (Uninscribed fragment from Cave of Letters).
Two different and complementary vibrational techniques were
used to investigate the samples, as described below.
3.1. Raman
Two kinds of Raman instruments were available, one to perform
confocal ␮-Raman, the other to collect FT-Raman spectra.
For confocal ␮-Raman measurements a Renishaw In-Via Reflex
Raman microscope was used, equipped with a Renishaw diode
laser at 785 nm (output power 300 mW). In order to avoid sam-
ple degradation, neutral filters, when necessary, were applied. The
backscattered light was dispersed by a 1200 line/mm grating and
the Raman signal was detected by a Peltier cooled (203 K) deep
depletion charge-coupled device (CCD, 578 × 384 pixel) optimized
for near-infrared and ultraviolet. Nominal spectral resolution was
about 3 cm−1 . The system, equipped with a Leica DMLM micro-
scope to focus the laser on the sample and a colour video camera,
allows for positioning the sample to select a specific region for the
investigation.
Spectral acquisitions (5–10 accumulations, 50 s each) were per-
formed with a 50 × objective. Under these conditions, the laser spot
measures about 20 ␮m2 .
For FT-Raman measurements a Bruker FT Raman spectropho-
tometer RFS 100/S was used, equipped with Ge-diode detector
cooled to liquid nitrogen temperature (77 K) and a near infrared
[YAG:Nd] laser operating at 1064 nm. Measurements were carried
out with the output power of 50–100 mW in the spectral range
100–3600 cm−1 with a spectral resolution of 4 cm−1 , 4000 scans
were co-added per spectrum.
3.2. Attenuated Total Reflection-Fourier Transform InfraRed
(ATR-FTIR)
Infrared spectra were acquired with a Nexus Nicolet interfer-
ometer, equipped with a KBr beam splitter and extended with a
ZnSe cell and a liquid nitrogen cooled MCT/A detector.
Measurements were performed in the 4000–650 cm−1 range at
a resolution of 8 cm−1 , averaging 300 acquisitions per sample.
24 M. Bicchieri et al. / Journal of Cultural Heritage 32 (2018) 22–29
Fig. 1. ATR-FTIR spectra of the mock-ups GB0 to GB4 prepared as described in
section 3.A. Shaded areas highlight the regions of interest for tannin detection.
4. Results and discussion
The mock-up samples (A in section 3) were especially man-
ufactured in order to reproduce ancient Oriental Jewish ritual
parchments. They were analyzed as well as modern and ancient
Western Jewish ritual parchments (B in section 3) to test whether
our techniques were suitable for their analysis and to determine
which band or bands combination could be useful for the charac-
terization of the historical samples (C in section 3).
The need of mock-up derives from the knowledge that each
parchment preparation, treatment and degradation can induce
structural modifications that affects the spectral features and must
be documented.
The main difference between Oriental and Western parchment
preparation consisted in the dehairing method: enzymatic for Ori-
ental and lime-based for Western [1,7–9]. Moreover, a finishing
treatment with tannin was supposed to be applied on ritual Jewish
parchments, as described in Talmud.
4.1. Mock-ups
ATR-FTIR spectra of all laboratory samples are shown in Fig. 1,
whereas micro-Raman and FT Raman spectra of the same samples
are plotted in fig. 2 A and B, respectively. Table 1 reports the ATR-
FTIR bands for all the laboratory samples with their attribution.
Fig. 2. Raman spectra of mock-ups. A. Confocal ␮-Raman spectra. B. FT-Raman spectra. The different preparation of mock-ups is described in section 3.A. Arrows identify
the peak related to tannin.
The ATR-FTIR GB0 spectrum shows the typical absorption bands
of the parchment [10,11].
In ATR-FTIR, the tannin peak is well visible in GB3 and GB4 sam-
ples, as well as in confocal ␮-Raman (Fig. 2A), as described below.
The tannic acid treatment is evidenced in ATR-FTIR by the peaks
in the range 1703–1727, at 1321 (GB3) or 1323 (GB4) and at 1190
(GB3) or 1199 (GB4) cm−1 that are well distinguishable from the
parchment spectra.
The bands in the range 1703–1727 cm−1 are attributable to
the carboxylic C = O stretching; the band around 1322 cm−1 is the
symmetric stretching of C-O(H) and that at 1190/1199 cm−1 is the
superimposition between the stretching of C-O-(C) and the bending
of C-H [12,13].
Even though the tannins most intense and characteristic ATR-
FTIR features are recognized in the 1500–1750 cm−1 range (C = C
ring quadrant stretching and C = O stretching of carboxyl) [14], in
the case of parchment substrate, they superimpose to the Amide I
and Amide II stretching modes, that lie in the same range [12,15,16].
The bands at 1115 cm−1 (not attributed, but possibly related
to some interaction between parchment and carbohydrates)
at around 1070 cm−1 (C-O-(C) asymmetric stretching in the
saccharide structure) could be indicative of the presence of polysac-
charides (gum Arabic and Acacia Raddiana) in the parchment
manufacture.
In Raman (Fig. 2A and B), GB0 sample shows the typical features
reported in literature for the parchment [23]. The treatment with
tannins is recognisable by the main distinctive peak at 1611 cm−1
related to the tannin ring C = C stretching, visible both in ␮- and
FT-Raman [24]. The spectral region 1300–1600 cm−1 involves the
aromatic ring vibrations: specifically, the 1500–1600 cm−1 range
involves the C-H in plane bending and the C = C ring vibrations in
which two opposite quadrants of the benzenic ring stretch, while
the other quadrants contract [14]. In the 1400–1500 cm−1 region,
the C = C ring vibration is owed to the stretching of one semicircle of
the ring, while the other contracts still together with C-H bending
[14]. The band around 1340 cm−1 could be attributable to C-H in
plane deformation.
The treatment with gum Arabic is not evidenced by Raman
spectroscopy, while the addition of Acacia Raddiana induces a
broadness of the bands in confocal ␮-Raman and does not modify
the spectrum of parchment when FT-Raman is applied.
Concerning the sample GB3, a better definition of tannin pres-
ence is obtained with confocal ␮-Raman measurements. The tannin
peak at 1611 cm−1 is present in both spectra, whereas the peak at
1710 cm−1 , typical of tannic acid, is not visible in FT-Raman. The
two peaks are marked with arrows in Fig. 3, where the spectrum of
GB0 is reported for comparison.
 M. Bicchieri et al. / Journal of Cultural Heritage 32 (2018) 22–29 25

Table 1
FTIR bands of the laboratory samples with their attribution.

GB0 GB1 (GB0 + gum Gum Arabic GB4 (GB0 + gum GB3 (GB0 + tannin) GB2 Possible assignments
Arabic) Arabic and tannin) (GB0 + acacia
raddiana)

3281, 3080 3287, 3079 3296 3260 NH stretching

2922 2927 2925 2930 2953, 2910 2921 CH2 asymmetric stretching
2860 2855 2841 2852 CH2 sym. stretch.
1725 1720, 1703 carboxylic C = O stretching of tannins
1660 (sh) 1660 sh 1665 sh C = O stretching Amide I ␣-helix
1639 1639 1642 1637 1640 C = O stretching Amide I ␤-sheet; C = O
asymmetric stretching acacia [17]
1616 1629 1616 C = O stretching
1550 1550 1550, 1533 1542, 1524, 1508 1555, 1543, NH bend coupled with CN stretching
1531, 1513 Amide II
1454 1451 1456 1459 1455 CH2 bending Amide II
1430–1350 broad 1455–1378
1444 1440 1446 scissoring of, O-C-H, C-C-H, CH2 , C-OH
[18]
1398 w 1400-1370 CH2 wagging
broad
1377 w 1372 OH scissoring
1338 w 1337 1343 sh 1343 sh 1339 CH2 rocking
1323 1321 C-O(H) symmetric stretching tannins
1315 w 1308 CH2 twisting
1280 1278 1283
1238 1237 1230 1239 NH bending Amide III
1203 1203 1203 C-H bending of parchment
1199 1190 C-O-(C) stretching of tannins coupled
with C-H bending of parchment
1160 1160 CH2 twisting in proline [19]
1142 1139 CH in plane bending
1115 1115 not attributed possibly related to
interaction between parchment and
carbohydrates
1083 1080 1089 1097 ring stretching of proline and
hydroxyproline [20]
1079 C-O stretching of secondary alcohols in
hydroxyproline

1072 1070 1070 C-O-C asymmetric stretching of the

saccharide structure (gum Arabic and
acacia)
1062 1055 1067 1066 1065 C-H out-of-plane-bending in proline
controllare [21]
1029 1038 1026 1033 1023 1028 C-C, C-OH, C-H ring and side group
vibrations [22]
1009 1013 1007 1009
971 982, 804 969 876 962, 895, 852, C-O and C-C stretching of the saccaride
815 structure/ring
726, 697 703, 679 777, 714, 668 752 scheletal

4.2. Modern and ancient Western Jewish parchments

ATR-FTIR and Raman spectroscopies were then applied to the

modern Western Jewish ritual samples S1 and S2 and to the ancient
S3 and S4 original samples.
Confocal ␮-Raman and FT Raman spectra of all the samples did
not evidence the presence of tannins but only the presence of the
applied salts, calcite in S1 and S3, alum in S2, whereas S4 sample
was affected by an intense fluorescence background that masked
the Raman signals. Probably, the used concentration of tannin was
too low to be detected.
On the other side, ATR-FTIR spectra were dominated by the sig-
nals of the salts (calcite and alum), but the tannate band at about
1743 cm−1 was visible. In Fig. 4 we report, as example, S1 and S2
ATR-FTIR spectra.

4.3. Selected fragments from Reed archive

Fig. 3. Confocal ␮-Raman and FT-Raman spectra of the tanned GB3 parchment in the
1200–1800 cm−1 region. GB0 spectrum is reported for comparison. Arrows identify
the peaks related to the tannin treatment.
4.3.1. Sample R-4Q1
ATR-FTIR measurements on R-4Q1 sample showed the presence
of tannate (1726–1610–1320 cm−1 ), gypsum (1622–1135 cm−1 )
and calcite (1400–874 cm−1 ), whereas FT-Raman spectra (Fig. 5)
26 M. Bicchieri et al. / Journal of Cultural Heritage 32 (2018) 22–29

Table 2
Gauss fit for Raman results on R-4Q1 in the salts region.

R-4Q1 sample fit in 588–656 cm−1 region

Peak Wavenumber Width Area Chi2 /DoF; R2

1 589 ± 1 4.1 ± 0.5 1.5 10−3 ± 0.3 10−3 Chi2 /DoF = 9.98
2 602 ± 1 16.1 ± 0.5 3.7 10−2 ± 0.1 10−2 10−10
3 621 ± 1 10.8 ± 0.4 4.3 10−2 ± 0.4 10−2 R2 = 0.999
4 634 ± 1 13.4 ± 2.3 3.1 10−2 ± 0.6 10−2
5 646 ± 1 8.5 ± 0.5 1.3 10−2 ± 0.3 10−2

R-4Q1 sample fit in 981–1018 cm−1 region

Peak Wavenumber Width Area Chi2 /DoF; R2

1 993 ± 1 7.8 ± 0.2 7.4 10−2 ± 1.6 10−3 Chi2 /DoF = 3.86 10−8
2 1006 ± 1 9.5 ± 0.1 1.2 10−1 ± 1.7 10−3 R2 = 0.997

Table 3
Position of peaks of different sulphates, after Gaussian fit of Raman spectra.

Sulphate of Peaks cm−1

Fig. 4. ATR-FTIR spectra of modern Western Jewish ritual samples S1 (manufactured
with calcite and oak gall) and S2 (manufactured with alum and oak gall). The figure Na 453 621 989 1101 1132
presents only the fingerprint region, useful for the attribution of salts and tannin, Al hydrate 605 992 1090
all marked by arrows. Al anhydrous 466 623 954 995 1101 1137
Al/K 616 974 989 1103 1131
Mg/Al 421 465 528 616 974 995 1074
NH4 /Al 458 522 621 994 1130
(Tschermigite)
NH4 /Al/Fe 445 461 623 997 1135
(Godovikovite)

Table 4
Attribution of salts on R-4Q1 sample.

4RQ1 peaks from fit cm−1 Standard peaks cm−1 Compound

993 994 (Rruff) Tschermigite

621 621 (Rruff)
1006 1007 (UCL database) Gypsum
1008 (Rruff, Italy cave)
1008 (Rruff, USA cave)

By comparison with the standards, it is possible to assume that

the sample R-4Q1 contains gypsum. The best fit was found with
Tschermigite NH4 Al(SO4 )2 ·12H2 O, as can be seen in Table 4.

Fig. 5. Spectra FT-Raman (top) of the non-tanned original R-4Q1 and m-Raman
(bottom) of R-4Q2 samples (Uninscribed fragments from Qumran, Cave 4).
did not evidence any treatment with tannin, but showed nar-
row peaks in the 588-656 cm−1 and 981–1020 cm−1 regions,
attributable to the salts described below. It was not possible to
collect spectra in confocal mode, due to an intense fluorescence
background.
As previously commented in the mock-up analysis, infrared
spectroscopy is more sensitive to the presence of tannins on the
parchment; on the contrary Raman is very useful for the identifi-
cation of salts.
A fit was performed in two regions: 588–656 cm−1 and
981–1020 cm−1 for the attribution of salts (Fig. 6). At wavenum-
bers lower than 588 cm−1 , the quality of the Raman signal was too
poor to perform the fitting. The results of the fitting are reported in
Table 2.
From the two fits, the following peaks were detected in the
sample: 588.8, 601.6, 620.7, 634.1 (sh), 645.7, 992.6, 1005.5,
1127.7 cm−1 .
Spectra of several sulphates, among them different ammo-
nium sulphates (Tschermigite, Godovikovite, hydrate ammonium
aluminium sulphate, anhydrous ammonium aluminium sulphate)
were analyzed with the same fitting method (Table 3) to obtain the
best attribution for R-4Q1 spectra.
4.3.2. Sample R-4Q2
ATR-FTIR and Raman measurements on R-4Q2 sample did not
reveal tannins. Salts on the surface were detected only by confocal
␮-Raman spectroscopy (Fig. 5).
Also in this case, a fit was performed in 1050–1150 cm−1 region
for the attribution of salts (Table 5 and Fig. 7).
By comparison with standards analyzed with the same fitting
method, it is possible to state the presence of calcite and suppose
the presence of dolomite and some iron carbonate (Table 6). The
presence of the latter could justify the broadening of the dolomite
peak and its shift to lower wavenumbers.
4.3.3. Sample R-4Q11
All measurements showed a parchment in a good state of con-
servation.
As usually, ATR-FTIR measurements evidenced the presence of
tannin, which peak was not visible by using FT-Raman.
Conversely, by confocal ␮-Raman the tannin peak around
1611 cm−1 was well evident (Fig. 8).
The different response of the two techniques is related to the
different sampling area.
The tannin peak on parchment is not very strong, due to the low
amount of tannin applied on the surface, and its band intensity is
reduced when the spectrum is averaged on a greater surface, as in
FT-Raman.
 M. Bicchieri et al. / Journal of Cultural Heritage 32 (2018) 22–29 27

Fig. 6. Deconvolution of the Raman spectrum of R-4Q1 (Uninscribed fragment from Qumran, Cave 4) in the 588–656 cm−1 region (left) and in the 981–1020 cm−1 region
(right). Black line: original spectrum; gray line: reconstructed spectrum after deconvolution; dotted line: peaks from fitting.

Table 5
Gauss fit of Raman spectra on R-4Q2 in the salt region.

Peak Wavenumber Width Area Chi2 /DoF; R2

−3 −3
1 1062 ± 1 6.3 ± 0.6 3.5 10 ± 0.3 10
2 1085 ± 1 8.5 ± 0.1 5.9 10−2 ± 1.2 10−3 Chi2 /DoF = 5.57 10−9
3 1096 ± 1 7.0 ± 0.6 2.7 10−2 ± 3.9 10−3 R2 = 0.998
4 1103 ± 1 7.3 ± 0.5 2.2 10−2 ± 3.13 10−3
5 1123 ± 1 8.1 ± 0.6 5.8 10−3 ± 0.3 10−3

Fig. 7. Deconvolution of the Raman spectrum of R-4Q2 in the 1050–1150 cm−1

region (Uninscribed fragment from Qumran, Cave 4). Black line: original spectrum;
Fig. 8. Confocal ␮-Raman and FT Raman spectra of R-4Q11 sample (Uninscribed
gray line: reconstructed spectrum after deconvolution; dotted line: peaks from fit-
fragment from Qumran, Cave 4) compared to GB0 mock-up. Confocal ␮-Raman
ting.
shows the treatment of the parchment with tannin (peak at 1611 cm−1 ), not revealed
by FT-Raman.

Table 6
Since in a previous work [25], the authors reported that the Attribution of salts on R-4Q2 sample.
tannin peak is well visible in confocal ␮-Raman, when a solution 4Q2 peaks Standard peaks compound
of tannic acid with concentration equal or greater than 10 g/L is
1085 1087 (Rruff) Calcite
applied on the surface, it is possible to assess that the R-4Q11 sam- 1085
ple had been prepared with a large amount of tannin. (archaeological
sample)
1096 1098 (Rruff) Dolomite
1091 (Rruff) Alumohydrocalcite
4.3.4. Samples R-M10 and R-A45 1085 (Rruff) CaAl2 (CO3 )
R-M10 and R-45 are uninscribed fragments respectively from 2 (OH)4 ·3H2 O

the Cave Murabba’at and the Cave of Letters.
In both archaeological samples analyzed with confocal ␮-
Raman, the presence of tannin was evident (Fig. 9), thus confirming
that the method of parchment manufacturing could comprehend
the use of tannins.
It was impossible to collect ATR-FTIR spectra, due to the high
roughness of the surface of the two samples.
Siderite (Fe CO3 )
4.3.5. Sample R-Q9 and R-M9
ATR-FTIR spectra of R-Q9 showed the presence of gum Ara-
bic, calcite and calcium oxalate, the last produced by fungal attack
[26–28] (Fig. 10), whereas ATR-FTIR spectra of R-M9 only showed
the presence of calcite and gum Arabic.
28 M. Bicchieri et al. / Journal of Cultural Heritage 32 (2018) 22–29
Fig. 9. Confocal ␮-Raman spectra of original tannate samples. Lower spectrum R-
M10 (Uninscribed fragment from Cave Murabba’at) and upper spectrum R-A45
(Uninscribed fragment from Cave of Letters).
Fig. 10. ATR-FTIR spectra of R-Q9 (Uninscribed fragment from unknown cave)
evidencing the formation of calcium oxalates on the surface and the presence of
calcite and gum Arabic. The arrows identify the peaks related to tannin and the
different salts.
No tannin treatments were visible on both samples.
Raman spectra of both samples were affected by a huge fluores-
cence background. It was only evident the collapse of the amide-
region bands, due to a strong degradation and the presence of cal-
cite.
5. Conclusions
In many cases FT-Raman and ␮-Raman give similar results. The
difference in the sampling areas of the two techniques allows for
a better characterization of salts and compounds present in low
amount, by using ␮-Raman.
It is, in fact, possible to point the laser exactly in the area to be
investigated; therefore the signal is not affected by the averaged
spectra arising from a larger area. Conversely, if the analyzed sam-
ple gives a fluorescence background, FT-Raman is more useful for
the general and indicative characterization.
Very interesting is the comparison between the two techniques
applied on tanned parchment: confocal ␮-Raman is more efficient
in evidencing the peak related to the presence of tannin (around
1611 cm−1 ) as can be seen in GB3, GB4 and R-4Q11 spectra. In
contrast, it was possible to detect tannin with FT-Raman only in
the case of GB3.
It is also important to compare Raman and IR: some features (i.e.
gum Arabic) are better visible in IR and the presence of tannins is
always detected by infrared spectroscopy.
By using the three techniques applied, it was possible to rec-
ognize different salts, as discussed in the text, and three different
kinds of preparation in the archaeological Dead Sea Scrolls samples:
• non-tanned (represented by R-4Q2, R-Q9 and R-M9);
• lightly tanned (represented by R-4Q1 and R-4Q11);
• tanned (represented by R-M10 and R-A45).
These experimental results partially modify the previous data
of Ronald Reed, who stated, at a first time, that all DSS were treated
with tannin, supporting the description available in Talmud. In a
second time, he revised his conclusions and reported that one vari-
ety of the parchment had never been tanned. We demonstrated
now, working on the same material, that there is a third kind of
manufacture, which use only very small amount of tannin.
Moreover, the collected data allowed to evidence the following
points:
• the spectral features obtained by Raman and IR spectroscopies
are modified by tanning treatments that are recognisable in both
techniques, with different sensitivity. The finishing treatments
of parchment with tannin were revealed not only in the labora-
tory samples, but also in ancient specimens originating from the
Judean Desert and in Western sacred Jewish parchments;
• not all the archaeological samples were manufactured as
described in Talmud;
• the salts found on the surface can derive from the degradation
processes or microbiological attacks (gypsum, oxalates), storage
environment before the discovery or from the manufacture.
The parchments subjected to a superficial treatment with tan-
nins become more similar to leather [29] and the treated layer can
modify its dimensions in a different way in respect to the non-
treated stratum when the environmental conditions change. The
different behaviour of the two layers can induce the delamination
of the sample, as verified in many ancient ritual Jewish documents.
References
[1] J.B. Poole, R. Reed, The preparation of leather and parchment by the Dead Sea
Scrolls community, Tech. Cult. 3 (1) (1962) 1–26.
[2] R. Reed, J.B. Poole, A study of some Dead Sea Scroll and leather fragments from
Cave 4 at Qumran part I – physical examination, Proc. Leeds Phil. Soc. 9 (1)
(1962) 1–13.
[3] R. Reed, J.B. Poole, A study of some Dead Sea Scroll and leather fragments from
Cave 4 at Qumran part II - chemical examination, Proc. Leeds Phil. Soc. 9 (6)
(1965) 171–182.
[4] J.B. Poole, The nature, origins and techniques of manufacture of those of the
Dead Sea scrolls which are made from animal skins, PhD thesis, University of
Leeds, 1960.
[5] R. Reed, Ancient skins, parchments and leathers, Seminar Press, New York,
1972.
[6] I. Rabin, G. Brooke, J. Hodgson, E. Pantos, J. Prag, The Ronald Reed archive at the
John Rylands University Library, 4, e-Preservation, 2007, pp. 9–12.
[7] M. Falkiewicz-Dulik, K. Janda, G. Wypych, Handbook of material biodegrada-
tion, biodeterioration and biostablization, second ed., Chemtech Publishing,
Toronto, 2015, pp. 190.
[8] M.R. Derrick, D. Stulik, J.M. Landry, Infrared Spectroscopy in Conservation Sci-
ence, The Getty Conservation Institute, Los Angeles, 1999, pp. 190.
[9] R.J. Forbes, Studies in ancient technology, V, E.J. Brill, Leiden, 1966, pp. 1–77.
[10] E. Mannucci, R. Pastorelli, G. Zerbi, C.E. Bottani, A. Facchini, Recovery of ancient
parchment: characterization by vibrational spectroscopy, J. Raman Spectrosc.
31 (2000) 1089–1097.
[11] M.T. Rosado, M.L.T.S. Duarte, R. Fausto, Vibrational spectra of acid and alkaline
glycine salts, Vib. Spectrosc. 16 (1998) 35–54.
[12] I. Mohammed-Ziegler, F. Billes, Vibrational spectroscopic calculations on pyro-
gallol and gallic acid, J. Mol. Struct.: Theochem 618 (2002) 259–265.
[13] N. Melniciuc Puica˘, A. Pui, M. Florescu, FTIR spectroscopy for the analysis of
vegetable tanned ancient leather, Europ. J. Sci. Theol. 2 (4) (2006) 49–53.
 M. Bicchieri et al. / Journal of Cultural Heritage 32 (2018) 22–29
[14] N.B. Colthup, L.H. Daly, S.E. Wiberley, Introduction to infrared and Raman spec-
troscopy, third ed., Academic Press Inc, San Diego, 1990.
[15] H. Ozgunay, Ö. Sarı, M. Tozan, Molecular investigation of Valonea Tannin, J. Am.
Leather Chem. Assoc. 102 (2007) 154–157.
[16] C. Chadefaux, A.-S. Le Hô, L. Bellot-Gurlet, I. Reiche, Curve-fitting micro-ATR-
FTIR studies of the amide I and II bands of type I collagen in archaeological bone
materials, e-PS 6 (2009) 129–137.
[17] S. Muthulakshmi, Investigations on the biopolymers Papain, Gum Aca-
cia, Gum Tragacanth and Gum Guar: physical and antimicrobial prop-
erties, in: Chapter 2, Structural and optical properties of biopoly-
mers, PhD Thesis, Manonmaniam Sundaranar University, 2013, pp. 21–46
[http://www.shodhganga.inflibnet.ac.in/handle/10603/36798].
[18] M. Ibrahim, M. Alaam, H. El-Haes, A.F. Jalbout, A. de Leon, Analysis of the struc-
ture and vibrationals spectra of glucose and fructose, Ecletica Quimica 31 (3)
(2006) 15–21.
[19] A. Barth, The infrared absorption of amino acid side chains, Progr. Biophys. &
Molec. Biol. 74 (2000) 141–173.
[20] Y. Sheena Marya, L. Ushakumarib, B. Harikumarc, H. Tresa Varghesed,
C. Yohannan Panicker, FT-IR, FT-Raman and SERS Spectra of L-Proline,
J. Iran. Chem. Soc. 6 (1) (2009) 138–144 [http://www.sdbs.db.aist.go.jp/
sdbs/cgi-bin/direct frame top.cgi. accessed 18 October 2017].
[21] R. Ramasamy, Vibrational spectroscopic studies of imidazole, Armenian J. Phys.
8 (1) (2015) 51–55.
[22] Mizi Fan, Dasong Dai, Biao Huang, Fourier Transform Infrared Spectroscopy
for Natural Fibres, in: S. Salih (Ed.), Fourier Transform - Materials Anal-
ysis. InTech, 2012, http://cdn.intechopen.com/pdfs/37067/InTech[HYPHEN]
Fourier transform infrared spectroscopy for natural fibres.pdf (accessed 18
October 2017).
29
[23] H.G.M. Edwards, D.W. Farwel, E.M. Newton, F.R. Perez, S.J. Villar, Application
of FT-Raman spectroscopy to the characterisation of parchment and vellum, I;
novel information for paleographic and historiated manuscript studies, Spec.
Acta A 57 (6) (2001) 1223–1230.
[24] Y.B. Byaruhanga, M.N. Emmambux, P.S. Belton, N. Wellner, K.G. Ng, J.R.N.
Taylor, Alteration of Kafirin and Kafirin film structure by heating with
microwave energy and Tannin complexation, J. Agric. Food Chem. 54 (12)
(2006) 4198–4207.
[25] M. Bicchieri, M. Monti, G. Piantanida, A. Sodo e, M.T. Tanasi, Inside
the parchment, 2008, http://www.ndt.net/search/docs.php3?MainSource=65
&rppoffset=120 (accessed 18 October 2017).
[26] F. Pinzari, M. Zotti, A. De Mico, P. Calvini, Biodegradation of inorganic com-
ponents in paper documents: formation of calcium oxalate crystals as a
consequence of Aspergillus terreus Thom growth, Int. Biodeter. Biodegr. 64
(2010) 499–505.
[27] C.J. McNamara, N. Konkol, R. Mitchell, Microbial Deterioration of Cultural Her-
itage Materials, in: R. Mitchell, J.-D. Gu (Eds.), Environmental Microbiology,
Wiley Hoboken, 2010, pp. 137–151.
[28] M. Bicchieri, M. Monti, G. Piantanida, A. Sodo, Non-destructive spectroscopic
investigation on historic Yemenite scriptorial fragments: evidence of different
degradation and recipes for iron tannic inks, Anal. Bioanal. Chem. 405 (8) (2013)
2713–2721.
[29] M. Bicchieri, M. Monti, G. Piantanida, F. Pinzari, A. Sodo, Non destructive spec-
troscopic characterization of parchment documents, Vib. Spectrosc. 55 (2011)
267–272.