Study of the interaction between collagen and naturalized and commercial dyes by Fourier transform infrared spectroscopy and thermogravimetric analysis

Article abstract of DOI:10.1016/j.dyepig.2015.01.012

The naturalized dyes (ND) and the traditional acid dyes (ADs) were compared by studying the different behavior during the leather dyeing process. NDs are glyconjugated compounds synthesized by the covalent union of a dye species with a natural sugar (e.g. lactose) able to confer water-soluble properties to the dye molecule as a whole. The interactions between the dyes and the leather proteins were studied by FT-IR spectroscopy and thermogravimetric (TG) analyses. The protein cross-linking of the dyed leather samples was investigated by studying the 1654/1690 cm^-1 peak height ratio and a deconvolution procedure of the amide I peak. The helix secondary structure was the predominant component of the leather proteins of the samples dyed with low concentrations of NDs (2%), while the β-sheets prevailed when leather samples were dyed with the traditional ADs and high concentrations of NDs (>5%). The data were discussed with respect to TG results.

Full text of DOI:10.1016/j.dyepig.2015.01.012

Dyes and Pigments 116 (2015) 65e73
Contents lists available at ScienceDirect
Dyes and Pigments
journal homepage: www.elsevier.com/locate/dyepig
Study of the interaction between collagen and naturalized and
commercial dyes by Fourier transform infrared spectroscopy and
thermogravimetric analysis
a
b
b
b, **
Davide Pellegrini , Massimo Corsi , Marco Bonanni , Roberto Bianchini
,
a
a, *
Alessandro D'Ulivo , Emilia Bramanti
National Research Council of Italy, C.N.R., Istituto di Chimica dei Composti Organo Metallici-ICCOM-UOS Pisa, Area di Ricerca, Via G. Moruzzi 1, 56124 Pisa,
a
Italy
b
Department of Chemistry “Ugo Schiff”, Via della Lastruccia 3-13, 50019 Sesto Fiorentino, Florence, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 25 September 2014
Received in revised form
The naturalized dyes (ND) and the traditional acid dyes (ADs) were compared by studying the different
behavior during the leather dyeing process. NDs are glyconjugated compounds synthesized by the co-
valent union of a dye species with a natural sugar (e.g. lactose) able to confer water-soluble properties to
the dye molecule as a whole. The interactions between the dyes and the leather proteins were studied by
FT-IR spectroscopy and thermogravimetric (TG) analyses. The protein cross-linking of the dyed leather
samples was investigated by studying the 1654/1690 cm peak height ratio and a deconvolution pro-
cedure of the amide I peak. The helix secondary structure was the predominant component of the leather
proteins of the samples dyed with low concentrations of NDs (2%), while the b-sheets prevailed when
10 January 2015
Accepted 12 January 2015
Available online 21 January 2015
ꢀ1
Keywords:
Dyes
FT-IR
leather samples were dyed with the traditional ADs and high concentrations of NDs (>5%). The data were
discussed with respect to TG results.
TGA
Cross-linking
Collagen
Secondary structure
© 2015 Elsevier Ltd. All rights reserved.
1. Introduction
Recently, we synthesized a new class of environmentally
friendly naturalized dyes (NDs) [4]. Briefly, a naturalized dye is a
The transformation of animal skin into leather is a complex
process encompassing several steps, among which leather dyeing
adds value to the material for market purposes [1]. Routinely, the
dyeing is carried out with sulfonated colorants at acidic pH, in order
to establish electrostatic interactions between the residual amino
groups of collagen and the sulfonic groups of the dyes [2]. This
process takes place in water and in the presence of a wide variety of
auxiliary chemicals: from salts to adjust the pH of dyeing baths, to
surfactants and mordants to aid dye penetration into leather and
enhance the overall fastness properties. Therefore, effluents from a
tannery include a broad spectrum of contaminants, impacting
heavily on the environment [3].
glyconjugated chemical, which is the covalent union of a dye spe-
cies (e.g., azo, anthraquinone, aniline type chromophore) with a
natural sugar, privileging lactose, able to impart remarkable water-
soluble properties to the dye molecule as a whole. Moreover, dyeing
processes involving NDs do not require the addition of chemical
auxiliaries [5] and the resulting fastness properties of the dyed
materials (e.g., polyester and nylon) are competitive with those
observed from commercial colorants [6]. On the basis of these
findings we prepared a second generation of NDs, in view of
replacing traditional acid dyes (ADs) to advance to a more eco-
sustainable leather industrial process. In particular, we achieved a
new chemical bridge linking the chromophore and lactose [5],
introducing a more stable amide moiety in place of an ester (Fig. 1)
[4].
This modification to the first reported structures generated NDs
with an amphoteric character, which is essential for leather dyeing
process [5]. Furthermore, we observed that chemically different
glyconjugated dyes of this new class were able to dye leather when
*
Corresponding author. Tel.: þ39 050 315 2293; fax: þ39 050 315 2555.
** Corresponding author. Tel.: þ39 055 4573486; fax: þ39 055 4574913.
E-mail addresses: roberto.bianchini@unifi.it (R. Bianchini), bramanti@pi.iccom.
cnr.it (E. Bramanti).
combined among them, giving
a
finished material with
http://dx.doi.org/10.1016/j.dyepig.2015.01.012
0143-7208/© 2015 Elsevier Ltd. All rights reserved.

66
D. Pellegrini et al. / Dyes and Pigments 116 (2015) 65e73
patented experimental procedures, elaborating commercial
Disperse Orange 30 (DO30), Disperse Red 202 (DR202), Disperse
Blue 27 (DB27) and Disperse Yellow 42 (DY42), respectively [5].
Formic acid was purchased from Sigma-Aldrich-Fluka (F0507, pu-
rity > 95%). All solutions were prepared with deionized water
(18.2 MOhm cm) using a Milli-Q system (Millipore, Bedford, MA,
USA).
0
2
.2. Synthesis of second generation ND: the (piperazin-6 -yl)lactose
moiety
The selective protection of the hydroxyl groups of lactose 1 [25]
0
0
generated the protected diol 2, whose 6 and 2 positions were
available for further elaboration. Regioselective tosylation of the
primary alcohol [26] was followed by an S 2 process using an
N
Fig. 1. Naturalized dyes.
excess of piperazine to obtain 4 in 62% overall yield for three steps
(Scheme 1).
Next, the piperazine derivative 4 was coupled to the dye 5,
which had been prepared conveniently from chromophore DO30
[5]. The free carboxylic acid moiety was activated with 4-(4,6-
homogeneous hue and fastness properties [6]. Elasticity, flexibility
and longitudinal/tensile strength characterize collagen-based ma-
terials depending on the degree of intermolecular cross-links be-
tween the collagen triple helical units: lengthwise strength is
enhanced by a parallel alignment of the fibrils (tendons), compli-
ance is obtained through random layered arrangements (skin),
flexibility is gained with laminated sheets, tensile strength is
improved by means of concentric fibrillar layers [7]. These prop-
erties have been studied experimentally but also through suc-
cessful computational materiomics studies [8e12]. While amino
acid side chains play an important role in self-association of
collagen helices and lateral packing of collagen helices, they are also
crucial in the binding of a wide variety of molecular species which
influence the microstructure and physical performance of the
collagen matrix [13,14].
dimethoxy-1,3,5-triazin-2-yl)-4-methylmorpholinium
chloride
(DMTMM) in mild conditions [27], recovering compound 6 in good
yield. Final deprotection of the acetonide moieties was carried out
in acid conditions, to restore the original structure of the lactose
portion within 7 (Scheme 2) [5].
Adduct 7 showed a significant solubility in water, even though
the piperazinil unit was present as a hydrochloride salt. A similar
strategy was followed to synthesize compound 8 and 9 (Fig. 2) [5].
A slightly different approach was adopted for compound 12. In
this case, DY42 10 was made to react with ethyl chloroacetate in the
presence of a catalytic amount of potassium iodide [28]. The
resulting ester derivative was hydrolyzed to acid 11, which was
converted to 12 following the route described previously (Scheme
3).
In this work, we investigated the interaction of second genera-
tion NDs 7e9 and 12 with leather at molecular level by Fourier
transform infrared spectroscopy (FT-IR) and thermogravimetric
analyses (TGA). IR spectroscopy is a well-established technique to
analyze the secondary structure of polypeptides and proteins
Thanks to this easy procedure to achieve the second generation
NDs, it is possible to process chemically different chromophores in
a reproducible manner, whenever a free carboxylic acid is available
for amide coupling [5].
[15e19]. The IR spectral data of polypeptides and proteins are
usually interpreted in terms of the vibrations of a structural repeat
unit, which give rise to nine characteristic IR absorption bands,
namely amide A, B and I-VII [17e19]. Among these, amide I and II
bands are the two most prominent absorptions of the protein
2.3. Leather dyeing
ꢀ1
backbone, with amide I (1700-1600 cm ) being the most sensitive
one of the spectral region. The amide I band is due almost entirely
to the carbonyl stretch vibrations of the peptide linkages and the
frequencies arising from each component of the complex absorp-
tion correlate closely to each secondary structural element of the
proteins [20]. Thus, second derivative spectra allow the identifica-
tion of the various secondary structures present in the protein [18]
and the curve fitting can be applied to calculate the contribution of
each component of the absorption band to the secondary structure
The operational procedure for leather dyeing was an adaptation
to laboratory scale of the daily large scale dyeing of a tannery.
Briefly, 0.2 g of chrome tanned leather specimen were put in a
plastic tube with a 2 mL aqueous solution of dye at C1 concentra-
tion (4 mg, 2% w/w) and stirred (magnetically) for one hour at
ꢁ
ꢁ
2
0 C. The dyeing bath was heated at 55 C and 0.4 mL of water and
[18,21e24].
The results were compared to those obtained from a group of
commercial ADs. An interaction model between the dyes and the
leather proteins was described, indicating that NDs interact better
than ADs with the polypeptide matrix.
2
. Experimental section
2.1. Materials and solutions
Chromium tanned leather samples and the commercial tradi-
tional ADs Acid Orange 37 (AO37), Acid Yellow 49 (AY49), Acid Red
49 (AR249) and Acid Blue 113 (AB113) were provided by BIO-
KIMICA. The NDs 7-9 and 12 were synthesized according to
2
0
Scheme 1. Synthesis of protected (piperazin-6 -yl)lactose.

D. Pellegrini et al. / Dyes and Pigments 116 (2015) 65e73
67
Scheme 2. Synthesis of piperazine-derived ND.
spectrophotometer. The illuminant was D65 and the light source
wavelengths were 360-740 nm. The diameter of the spot was 5 mm
ꢁ
and the light incident angle was 10 . The specular component
included (SCI) and specular component excluded (SCE) coordinates
were acquired simultaneously.
2.5. FT-IR spectroscopy
Fig. 2. Second generation NDs 8 from DR202 and 9 from DB27.
Infrared spectra were recorded on a PerkineElmer Spectrum
One FTIR spectrophotometer, equipped with a universal ATR
accessory and a TGS detector. After recording the background
spectrum, sample spectra were obtained on different area of blank
and leather samples. For each sample, 128 interferograms were
recorded (in order to obtain a suitable S/N ratio), averaged and
Fourier-transformed to produce a spectrum with a nominal reso-
lution of 4 cm . The spectra were run and processed by means of
the PerkineElmer spectrum software and a written-in house Lab-
VIEW program for peak fitting, respectively. Control analyses were
performed on leather specimens before entering into the dyeing
4
mg of formic acid were added. The sample was stirred for 30 min
and then rinsed with 4 mL of water for three times and dried at
room temperature. In the case of NDs we performed also another
set of experiments increasing the amount of dye up to C2 (5% w/w),
C3 (10% w/w) and C4 (15% w/w) concentrations. It has to be high-
lighted that the w/w percentage of an ADs chromophore is
approximate, due to the presence of additives in commercial
products [2]. Instead, NDs are chemically pure products and the
percentage refers to the actual weighted amount of dye.
ꢀ1
Table 1
2.4. Colorimetry
L*a*b* coordinate values for both SCI and SCE configurations of all dyed leather
specimens.
The final color of the leather samples were characterized by
colorimetry [29]. The three coordinates of the CIELAB color space
Dye
%(w/w)
SCI
SCE
were measured by using a CM2600d KONICA MINOLTA portable
L*
a*
25.73
b*
39.27
L*
a*
25.65
31.7
36.03
36.37
39.59
39
37.35
36.16
ꢀ13.29
ꢀ10.86
ꢀ9.45
ꢀ7.58
ꢀ7.56
ꢀ7.23
ꢀ5.98
ꢀ4.61
32.87
48.62
3.79
b*
39.22
44.02
45.08
44.9
7
C1
C2
C3
C4
C1
C2
C3
C4
C1
C2
C3
C4
C1
C2
C3
C4
C1
62.92
57.97
53.6
53.28
48.11
41.18
39.59
37.25
53.56
49.7
44.54
40.93
81.47
80.16
78.81
77.84
55.68
52.46
32.12
78.02
82.38
85.85
62.9
31.73
36.09
36.39
39.65
38.98
37.27
36.1
44.06
45.14
44.83
13.66
14.73
14.39
14.53
ꢀ22.35
ꢀ25.83
ꢀ25.34
ꢀ24.93
25.79
40.23
49.71
54
57.91
53.59
53.21
48.08
41.06
39.45
37.11
53.4
49.59
44.49
40.8
81.35
80.04
78.7
8
9
13.64
14.74
14.43
14.57
ꢀ22.39
ꢀ25.87
ꢀ25.37
ꢀ25.01
25.75
40.19
49.68
54.02
40.19
11.56
22.21
57.18
1.37
ꢀ13.29
ꢀ10.88
ꢀ9.46
ꢀ7.58
ꢀ7.6
12
ꢀ7.25
ꢀ6
ꢀ4.62
32.81
48.56
3.76
77.7
AO37
40.07
11.48
ꢀ22.11
57.24
1.34
55.56
52.33
31.96
77.85
82.21
82.68
AR249
AB113
AY49
blank-1
blank-2
ꢀ10.69
ꢀ4.39
ꢀ4.27
ꢀ10.62
ꢀ4.36
ꢀ4.25
/
/
0.39
0.43
Scheme 3. Synthesis of an aniline type second generation ND 12 from DY42.

68
D. Pellegrini et al. / Dyes and Pigments 116 (2015) 65e73
process (blank-1) and subjected to dye conditions in the absence of
the colorants (blank-2).
2.6. Thermogravimetric (TG) analyses
TG analyses were carried out on an EXSTAR series TGA/DTA7200
apparatus, with nitrogen as purging gas (200 mL/min). Each sample
5-10 mg) was placed in an alumina sample pan and the run was
(
ꢁ
ꢁ
carried out at a standard rate of 10 C/min from 30 to 700 C in the
presence of nitrogen.
3
. Results and discussion
3.1. Dyeing efficiency
The leather dyeing efficiency of the NDs was compared to that of
the traditional ADs. The dyeing baths at the end of the dyeing
procedure were almost colorless (the absorbance decreased of 95%
after the first rinsing step) when ADs were used, indicating that the
leather samples had retained the dyes very efficiently. In the case of
NDs the dyeing baths and the three rinsing solutions were colored,
with a decrease of the absorbance of about 60% after the first
rinsing step (data not shown for brevity). The final colors of the
leather samples dyed with NDs and ADs were characterized by the
L*a*b* coordinate values for both SCI and SCE configurations
(Table 1).
Leather samples dyed with NDs showed a decrease in the
lightness and an increase in the colour saturation as the amount of
the dye increased from C1 to C4. Furthermore, the sections of the
leather samples dyed with NDs were completely colored, indicating
their ability to penetrate across the samples. On the contrary the
ADs did not penetrate the leather, as shown by the absence of
colour in the central region of the sample sections (Fig. 3).
Fig. 4. Comparison of representative FT-IR spectra of blank-1, blank-2, ND 7 and AO37
in the 1800e900 cm region.
ꢀ
1
stretching and NH bending) of the leather proteins. Absorption
ꢀ1
features in the fingerprint region from 1451 to 1000 cm were
attributed to CH wagging, CH deformation, CeN stretching and
The increasing concentration of ADs from C1 to C2 did not
improve dye penetration. This result was likely due to the poor
mechanical stir of the leather samples in the plastic tube, compared
to the rolling and tumbling effects on whole box calf type leather
2
3
CeOH stretching of leather proteins [31] and the contribution of
dye absorptions, which is small because of the low dye concen-
tration. The amide region shows significant differences.
2
specimens (4-6 m wide) during the large scale operations in tan-
A careful investigation of specific absorption bands is funda-
mental to obtain useful information about the interaction of dyes
with leather proteins. In particular, FT-IR spectroscopy has been
widely employed to study the collagen cross-linking in bone and
nery drums (50-60 m3 volume, running at 20% capacity) [30].
However, it suggests that NDs penetrate across the leather sample
better than ADs and also in mild stirring conditions.
ꢀ1
cartilage samples through the analysis of the 1660/1690 cm
absorbance (or peak area) ratio, which increases as the collagen
cross-linking increases [32e34]. Intermolecular cross-linking is
responsible for the mechanical strength of collagen fibrils. In vivo
intermolecular cross-links are due to the formation of covalent
bonds from aldehydes generated enzymatically from lysyl and
hydroxylisyl sidechains by lysyl oxidase [35]. Cross-linking process
is also due to the reaction of reducing sugars (e.g. glucose) which is
strictly related in vivo to aging processes [36]. The nonenzymatic
glycosylation of certain lysine and hydroxylysine residues is the
first step in a series of reactions known as the Maillard reaction
3
.2. FTIR spectroscopic analysis of leather samples dyed with NDs
and ADs
Leather is composed by structural proteins (88% collagen, 6%
keratin, 0.9% elastin) and non-structural proteins (3% albumins and
globulins, 2.1% mucins and mucoids). Collagen is the most abun-
dant protein of leather and therefore, it is likely to assume that FTIR
spectra of leather mainly depend on collagen absorptions.
Fig. 4 shows the comparison of representative FT-IR spectra of
ꢀ
1
blank-1, blank-2, ND 7 and AO37 in the 1800e900 cm region.
ꢀ
1
[37], and it involves the nonenzymatic condensation of a free amino
The strong absorption bands at 1650 and 1550 cm
were
group on a sugar aldehyde or ketone. The resulting Schiff base is
attributed to the amide I (C]O stretching) and amide II (CN
Fig. 3. Sections of the leather samples dyed with NDs (A) and ADs (B).

D. Pellegrini et al. / Dyes and Pigments 116 (2015) 65e73
69
likely to undergo a rearrangement to form a fairly stable ketoamine
called the Amadori product [38].
Barth et al. in a study on bone collagen reported that the sup-
pression of plasticity from fibrillar sliding gave an increase in the
samples treated with ADs were explored at C1, while the samples
treated with NDs were explored at C1 and C4. Each single
component of amide I was assigned according to the literature
ꢀ1
ꢀ1
[7,20,39,41,42], namely ca. 1680 cm to turns, 1690 cm to anti-
ꢀ
1
ꢀ1
1660/1690 ratio and it has been primarily attributed to an increase
parallel
b
b-sheets, 1656 cm to helix, 1630 cm to intramolecular
ꢀ
1
in the concentration of the nonenzymatic (AGE) cross-links. The
decrease in the 1660/1690 ratio is more explicitly related to
changes in enzymatic cross-links [39].
-sheets, 1610 cm to intermolecular b-sheets. The peak fitting
analysis showed that all the samples had almost the same sec-
ondary structure components, but the different percentages re-
flected the different type of interaction between the dyes and the
leather proteins (Table 2).
The helix is the predominant component of native collagen
secondary structure. Modifications observed in the component
percentages of leather samples suggested that major changes of
collagen in leather samples treated with NDs at C1 were the in-
It is quite clear that the collagen structural changes and the
eventual cross-linking responsible for the spectral changes
observed in dyed leather samples are not due to enzymatic pro-
cesses, but to the interaction/reaction with dyes and/or with formic
acid employed in the dyeing procedure.
ꢀ
1
Fig. 5 shows the values of the 1654/1690 cm absorbance ratio
calculated from the spectra of the dyed leather samples
investigated.
crease of helix structure and the decrease of b-sheets. Helix per-
centage in NDs at C1 dyed samples was indeed 48 ± 4%, not
statistically different from the blank-2 (47%). In samples dyed with
traditional ADs or NDs at C4 the helix percentage was instead
38 ± 4% and 31 ± 8%, respectively, comparable with the helix per-
centage found in blank-1 (39%). Thus, the structure of the leather
proteins (basically collagen) in samples dyed with NDs at C1 was
richer in helices, while in those treated with ADs at C1 and with
ꢀ1
The 1654/1690 cm absorbance ratio was found significantly
higher for the leather samples dyed with NDs at C1 and for the
blank-2 sample. The leather samples dyed with the ADs at the same
concentration gave an absorbance ratio similar to that of the
chrome tanned leather (blank-1).
By increasing the concentration of NDs, the ratio value
decreased, reaching the value calculated from the FT-IR spectrum of
the blank-1 (weak chrome-tanned leather) and the leather samples
dyed with ADs (Fig. 6).
These spectral changes may be the result of significant structural
changes of collagen, better discussed in the next paragraph.
NDs at C4 had a higher percentage of
b sheets.
Fig. 8 shows, for each dyed leather sample, the ratio of the
ꢀ1
component percentage at 1660 cm
assigned to helix and at
1690 cm assigned to turns and antiparallel sheets, analogous to
the ratio of the absorbance values at 1654 and 1690 cm
ꢀ
1
b
ꢀ
1
.
The results calculated on the basis of the Gaussian component
percentages confirmed the previous results calculated on the basis
of the ratio of absorbance values of the amide I band. The single
component analysis evidenced significant conformational changes
3
.3. Amide I peak fitting
The interaction of NDs and ADs with leather proteins and the
ꢀ1
ꢀ1
and significant variations of the 1660/1690 cm absorbance ratio.
It is interesting to note that blank-1 is a weak-chrome tanned
leather specimen before entering into the dyeing process; blank-2
is a weak-chrome tanned leather specimen processed with the
dyeing procedure in the absence of colorants. The blank-2 leather
spectral differences observed in the 1654/1690 cm ratio of Amide
I band of FT-IR spectrum, were studied more in detail applying a
peak fitting analysis to the vibrational C]O stretching frequencies
associated to the amide I (Fig. 7).
As mentioned above, different secondary structures of proteins
contribute to the overall stretching of the carbonyl double bond and
a second derivative spectrum analysis provides information about
the number and frequencies of the components. Secondary struc-
ture motifs were estimated by expressing the amplitude of the
bands, assigned to each structure as a fraction of the total sum of
the amplitudes of the amide I components [40]. The leather
ꢀ1
sample showed higher helix percentages and 1660/1690 cm
percentage components ratio values comparable with those ob-
tained in leather samples dyed with NDs at C1. These structural
ꢁ
changes can be likely due to the treatment at 55 C with
3
ꢀ
2
.63 ꢂ 10 M (1.67 g/L, i.e. 0.17% w/w) formic acid, conditions
commonly employed to fix the color without compromising the
Fig. 5. Absorbance ratio (1654/1690 cm 1) from the FT-IR spectra of leather samples dyed with NDs and Ads at C1 concentration. The dotted lines represent the ratio values of the
ꢀ
blank-1 and blank-2 samples.

70
D. Pellegrini et al. / Dyes and Pigments 116 (2015) 65e73
Fig. 6. Comparison of the absorbance ratio values (1654/1690 cmꢀ1) between leather samples dyed with NDs at C1-C4 and ADs at C1.
collagen structural integrity. The weak acid conditions of leather
pH < 3.5, isoelectric point of collagen) are mandatory in order to
collagen. The results obtained analyzing the blank-2 sample can be
explained assuming the reactions of formic acid analogous to those
reported for succinic acid.
(
favorite the interaction of sulphonic groups of ADs with protonated
amino groups of collagen [43]. We can hypothesize that in these
conditions formic acid may act as a cross-linker for collagen, as well
as succinic acid [44]. In the case of succinic acid the crosslinking is
due to the ion interaction between the carboxylate groups and the
protonated amine group of the collagen. The crosslinking between
succinic acid and collagen is also due to multiple intermolecular
hydrogen bonding between the acid and the lysine residue of the
Thus, all the conformational changes observed may be induced
by the primary interaction with colorants and/or by the formation
of cross-links among collagen fibers [45] induced by treatment
with formic acid and/or by colorants.
ꢀ1
The samples dyed with NDs at C1 had 1660/1690 cm absor-
bance ratio higher than those dyed with ADs at the same concen-
ꢀ1
tration and with NDs at C4. The 1660/1690 cm absorbance ratio of
Fig. 7. Representative FT-IR spectra in the amid I region of (A) blank-2, (B) C1 12, (C) C4 12, (D) C1 AY49, compared with the predicted spectra (bold dotted line). The second
derivative spectra (dashed line) and the single Gaussian components are also shown.

D. Pellegrini et al. / Dyes and Pigments 116 (2015) 65e73
71
Table 2
the leather samples dyed with 8, 9 and 12 at C4 was even lower
than that observed for traditional ADs.
Results of the deconvolution procedure applied to the amide I region of FT-IR spectra
of leather samples dyed with NDs and ADs.
Two phenomena may contribute to high cross-linking and
higher helix percentage observed in leather samples dyed with NDs
at C1: (i) the treatment with formic acid, (ii) the presence of the
lactose unit in the dye structure. The protonated structure of the
piperazine bridge of NDs (Scheme 2) and the contemporary pres-
ence of the polyhydroxylated structure of lactose may favor the
hydrogen bonding within the amino acid residues of collagen,
contributing to the stabilization of helix structure and to the in-
crease of cross-linking.
However, leather samples dyed with higher dye concentrations
up to C4 had lower helix and higher b-sheet percentages. This can
be likely due to the stacking interactions between chromophore
rings in excess [46] that surround collagen fibers, thus avoiding the
intermolecular covalent crosslinking and stretching the protein
structure because of their steric hindrance.
Frequency/cm 1
(
ꢀ
Assignment
%
bandwidth/cmꢀ1)
Blank-1
1630 (34)
1658 (34)
1610 (19)
1682 (43)
intramolecular
Helix
intermolecular
turns
b
-sheets
-sheets
42
39
5
b
13
Blank-2
1
1
1
1
7
1
1
1
8
1
1
1
1
9
1
1
1
1
1
1
1
1
1
630 (31)
659 (38)
611 (20)
intramolecular
Helix
intermolecular
b
-sheets
-sheets
37
47
6
b
693 (34)
Antiparallel
b
-sheets
9
a
632 (30)
659 (37)
intramolecular
Helix
Antiparallel b-sheets
b-sheets
36
54
11
692 (34)
a
629 (33)
659 (37)
610 (18)
intramolecular
Helix
intermolecular
b
-sheets
-sheets
40
46
4
3.4. Thermogravimetric (TG) analyses
b
691 (34)
Antiparallel
b
-sheets
10
The thermal stability of the leather samples dyed with NDs and
ADs was investigated by TGA. Fig. 9 shows a representative ther-
mogram of a leather sample dyed with 12 at C1.
a
628 (31)
609 (22)
657 (40)
692 (35)
intramolecular
intermolecular
Helix
b
b
-sheets
-sheets
36
10
45
8
The trends of the TG% and DTG were found to be not signifi-
cantly different for all the samples analyzed. The first weight loss
corresponded to the loss of water, while the second weight loss
corresponds to the polypeptide chain thermal decomposition.
Table 3 summarizes the TGA parameters found for all the samples
analyzed.
Antiparallel b-sheets
a
2
629 (33)
659 (37)
611 (18)
intramolecular
Helix
intermolecular
b
-sheets
-sheets
39
46
5
b
691 (34)
Antiparallel
b
-sheets
9
a
AO37
Fig. 10 shows the trend of the T02 values for all the samples
analyzed.
1
1
1
1
630 (38)
660 (34)
609 (17)
intramolecular
Helix
intermolecular
b-sheets
45
39
4
The thermal stability of the leather samples was found to be
dependent on the class and the percentage amount of dyes used in
the dyeing procedure. The leather samples dyed with NDs at C1 had
b
-sheets
689 (35)
Antiparallel
b
-sheets
11
a
AR249
1
1
1
1
1
629 (34)
658 (38)
603 (26)
684 (24)
intramolecular
Helix
intermolecular
turns
b
-sheets
-sheets
34
38
16
6
2
ꢁ
higher T0 values (338 ± 1 C) which corresponded to higher
thermal stability and were not statistically different from the blank-
b
ꢁ
2 (338 C). These results suggest that the treatment with formic
ꢁ
698 (22)
Antiparallel
b
-sheets
5
acid at 55 C increases the thermal stability of collagen fibers by
a
AB113
itself and low concentrations of NDs (C1) do not interfere with this
stabilization. The higher thermal stability of collagen fibers due to
treatment with formic acid found in blank-2 with respect to the
blank-1 sample is compatible with the hypothesis that formic acid
favors the cross linking process [44].
1
1
1
1
629 (35)
659 (36)
610 (18)
intramolecular
Helix
intermolecular
b-sheets
42
43
4
b
-sheets
690 (35)
Antiparallel
b
-sheets
10
a
AY49
1
1
1
1
7
1
1
1
1
8
1
1
1
1
1
9
1
1
1
1
1
1
1
1
1
635 (36)
601 (42)
661 (31)
intramolecular
intermolecular
Helix
b
b
-sheets
-sheets
36
21
32
11
The correlation between the thermal stability of the samples
ꢀ1
analyzed and the higher values of the 1660/1690 cm ratio found
in the FTIR study seems to support the correlation of this ratio with
crosslinked structures, in agreement with the literature data
687 (37)
Antiparallel b-sheets and turns
b
631 (32)
658 (35)
601 (35)
intramolecular
helix
intermolecular
turns
b
-sheets
-sheets
34
38
18
10
[32e34].
The leather samples dyed with ADs at C1 and NDs at C4 had
b
2
ꢁ
ꢁ
lower values of T0 (335 ± 1 C and 332 ± 2 C, respectively), with
values comparable (ADs) or even lower (8, 9 and 12 NDs showed
686 (38)
b
2
2
ꢁ
633 (32)
659 (33)
609 (37)
679 (31)
intramolecular
Helix
intermolecular
turns
b
-sheets
-sheets
30
35
21
9
the lowest T0 value) than the T0 found in blank-1 (334 C). In all
these cases it seems that the presence of the dyes interferes with
the mechanism of formic acid-induced cross linking. This interfer-
ence could be due to a “shielding effect” of stacked dye chromo-
phores around collagen fibers. These TG results agreed, indeed,
b
697 (27)
Antiparallel
b
-sheets
5
b
631 (50)
662 (29)
609 (15)
681 (58)
intramolecular
Helix
intermolecular
turns
b
-sheets
-sheets
55
19
4
with data obtained from FT-IR analysis. Higher values of the 1660/
ꢀ1
1
690 cm ratio evidenced in the leather samples dyed with NDs at
b
C1 samples suggested a more cross-linked structure of the samples,
compatible with higher thermal stability. Covalent bonds between
collagen helices stabilize, indeed, the material. Leather samples
dyed with ADs and NDs at C2-C4 showed lower thermal stability
21
b
2
633 (45)
662 (31)
613 (15)
687 (44)
intramolecular
Helix
intermolecular
b
-sheets
-sheets
52
30
3
b
ꢀ1
and corresponding lower values of the 1660/1690 cm ratio, a
decrease in helices and higher percentages of -sheets. Stacking
interactions between dye chromophores and -sheets structures of
Antiparallel
b
-sheets and turns
16
b
a
C1.
C4.
b
b

72
D. Pellegrini et al. / Dyes and Pigments 116 (2015) 65e73
Table 3
Experimental temperatures and weight loss percent of thermal degradation steps of
leather samples dyed with NDs and ADs.
1
2
Sample
T
0
T
ONSET
T
0
% Final weight
Blank-1
Blank-2
77.4
72.7
302.2
299.9
334.3
338.0
22.3
20.3
a
7
8
9
1
76.6
74.1
70.6
74.0
298.2
299.1
300.6
297.7
337.0
339.0
339.5
337.6
21.6
21.6
20.9
20.3
a
a
2a
AO37a
AR249
AB113
75.4
73.0
76.1
71.9
302.0
299.1
299.2
300.6
335.2
336.0
335.4
333.5
22.1
21.4
22.4
20.4
a
a
a
AY49
7b
77.7
79.0
72.7
76.5
297.4
299.9
299.3
299.5
334.6
330.6
330.9
331.5
22.5
22.1
22.7
22.6
b
8
b
9
1
2b
Fig. 8. 1660/1690 cmꢀ1 percentage components ratio calculated from FT-IR amide I
peak fitting of all leather samples dyed NDs at C1 and C4 and with ADs at C1. The
dashed lines correspond to the ratio values for the blank-1 and blank-2 samples.
a
C1.
C4.
b
aldehyde or ketone [37,38] with free amino groups, responsible for
the cross-linking. On the other hand, the hydroxyl groups of lactose
may interact by H-bonding with specific amino acid residues of
collagen, contributing to the stabilization of helix structure.
In samples dyed with traditional ADs at C1 and NDs at higher
collagen likely avoid the intermolecular covalent crosslinking and
may be responsible for the lower stability of these samples.
4
. Conclusions
concentrations (particularly at C4, 15%),
b-sheet structures were
predominant and the spectra were characterized by lower values of
the 1660/1690 cm ratio.
Traditional ADs, instead, interact with collagen primarily by
electrostatic forces between chromophore sulphonated groups and
the protonated amino groups of amino acid side chains [2].
FT-IR and TG analysis were useful to investigate the interaction
ꢀ1
of NDs, a new and eco-friendly class of dyes, and traditional ADs
with the proteins of chrome tanned leather at molecular level. The
NDs were found to exhibit a higher penetrating capacity into the
leather compared to that of traditional ADs, since they were capable
to impregnate the leather sample in mild agitation conditions. The
FT-IR analysis of amide I band, both in terms of the absorbance
Furthermore, the additional
p-stacking [47] occurring across the
AD molecules, exclude the hydrogen bonding cross-linkage,
essential for the stabilization of helix secondary structures. The
excess of NDs (up to C4) causes the inhibition of cross-linking, as
ꢀ1
ratios at two different wavelengths (the 1654/1690 cm absor-
bance ratio) and the analysis of the single components found by
peak fitting evidenced that the leather proteins in samples dyed
with NDs at C1 (2%) were characterized by higher percentage of
well, lower helix and higher
due to the stacking interactions between the excess of chromo-
phore rings that surround collagen fibers, (i) stabilizing the -sheet
b-sheets percentages. This can be likely
b
ꢀ1
helices and higher values of the 1660/1690 cm ratio, analogous to
the values found in blank-2 leather sample. We hypothesized that
all these conformational changes observed may be induced by the
primary interaction with colorants and/or by the formation of
cross-links among collagen fibers induced by the treatment with
formic acid and/or by colorants.
structures and (ii) avoiding the intermolecular covalent
crosslinking.
The thermoanalytical study confirmed these results and
matched with this hypothesis, revealing a higher thermal stability
for the leather samples dyed with NDs at C1 concentration and
lower thermal stability in leather samples dyed with ADs and NDs
at C4 concentration.
We hypothesized that the lactose unit in the NDs structure fa-
vors, by one hand, the nonenzymatic condensation of the sugar
Fig. 10. Trend of the T02 values for the leather samples dyed with NDs at C1 and C4
2
Fig. 9. Thermogravimetric curve (left axis) and its derivative (right axis) of leather
sample dyed with 12 at C1, performed under nitrogen flow at 10 C/min heating rate.
and with ADs at C1. The dashed lines correspond to the T0 values for the blank-1 and
blank-2 samples.
ꢁ

D. Pellegrini et al. / Dyes and Pigments 116 (2015) 65e73
73
In principle, NDs may be competitive with the traditional AD
because of their eco-friendly properties, because of their better
penetrating capacity and because their employment at low con-
centrations (about 2%) give more cross-linked structures. This last
characteristic may provide better quality features to the leather in
terms of resistance and flexibility.
Currently, studies for scale-up dyeing at industrial level with
NDs are in progress [48], taking into account the recently displayed
biodegradability properties of this new class of colorants in the
presence of common bacteria strains (e.g., Escherichia coli). This
would offer the opportunity to treat dyeing effluents in an eco-
friendly manner and re-use the water for further dyeing cycles,
cutting the costs associated to water management [49].
[20] Kong J, Yu S. Fourier transform infrared spectroscopic analysis of protein
secondary structures. Acta Biochimica Biophysica Sinica 2007;39(8):549e59.
[
21] Venyaminov SY, Kalnin NN. Quantitative IR spectrophotometry of peptide
compounds in water (H O) solution. I. Spectral parameters of amino acid
2
residue absorption bands. Biopolymers 1990;30:1243e57.
[22] Venyaminov SY, Kalnin NN. Quantitative IR spectrophotometry of peptide
2
compounds in water (H O) solution. II. Amide absorption bands of poly-
peptide and fibrous protein in alpha, beta and random coil conformations.
Biopolymers 1990;30:1259e71.
[23] Kalnin NN, Baikalov IA, Venyaminov SY. Quantitative IR spectrophotometry of
peptide compunds in water (H
2
O) solution. III. Estimation of the protein
secondary structure. Biopolymers 1990;30:1273e80.
[24] Dong A, Huang P, Caughey WS. Protein secondary structure in water from
second derivative amide I infrared spectra. Biochemistry 1990;29:3303e8.
[
25] Barili PL, Catelani G, D'Andrea F, De Rensis F, Falcini P. Improved preparation
0
0
0
of 2,3:5,6:3 ,4 -tri-O-isopropylidenelactose dimethyl acetal and its 6 -O-(1-
methoxy-1-methylethyl) derivative. Carbohydr Res 1997;298:75.
26] Kabalka GW, Varma M, Varma RS, Srivastava PC, Knapp Jr FF. The tosylation of
alcohols. J Org Chem 1986;51:2386.
[
Acknowledgments
[27] Kunishima M, Kawachi C, Iwasaki F, Terao K, Tani S. Synthesis and charac-
terization of 4-(4,6-dimethoxy-1,3,5-triazin-2-yl)-4-methylmorpholinium
chloride (DMT-MM). Tetrahedron Lett 1999;40:5327.
This work has been financially supported by the European
Project Lifeþ 2012 ENV/IT/352-“BIONAD”. The authors would like
to thank Dr. Stefano Legnaioli, Dr Giulia Lorenzetti and Dr. Ema-
nuela Grifoni for the colorimetric analysis.
[28] Baughman TW, Sworen JC, Wagener KB. The facile preparation of alkenyl
metathesis synthons. Tetrahedron 2004;60:10943.
[29] Schanda J. Colorimetry. Understanding the CIE system. Wiley Interscience;
2007.
[
[
30] www.dose.eu/products.php.
31] Wang Q, Sanad W, Miller LM, Voigt A, Klingel K, Kandolf R, et al. Infrared
imaging of compositional changes in inflammatory cardiomyopathy. Vib
Spectrosc 2005;38(1e2):217e22.
References
[
32] de Campos Vidal B, Mello MLS. Collagen type I amide I band infrared spec-
[
[
1] Haroun AA, Mansour HF. Effect of cationisation on reactive printing of leather
and wool. Dyes Pigments 2007;72:80e7.
2] Püntener AG, Pedrazzi R, Clausen T, Hunger K, Bauer W, Filosa M, et al.
Nontextile dyeing. industrial dyes. Wiley-VCH Verlag GmbH & Co. KGaA;
troscopy. Micron 2011;42(3):283e9.
[33] Twardowski J, Anzenbacher P, Masson MR. Raman and IR spectroscopy in
biology and biochemistry. Ellis Horwood; 1994.
[34] Blank RD, Baldini TH, Kaufman M, Bailey S, Gupta R, Yershov Y, et al. Spec-
troscopically determined collagen Pyr/deH-DHLNL cross-link ratio and cris-
tallinity indices differ markedly in recombinant congenic mice with divergent
calculated bone tissue strenght. Connect Tissue Res 2003;44:134e42.
[35] Eyre DR, Wu JJ. Collagen cross-links. Top Curr Chem 2005;247:207e29.
[36] Last JA, Armstrong LG, Reiser KM. Biosynthesis of collagen crosslinks. Int J
Biochem 1990;22(6):559e64.
[37] Maillard LC. Action of amino acids on sugars. Formation of melanoidins in a
methodical way. Compt Rend 1912;154:66.
[38] Amadori M. Um
1925;6:337.
[39] Barth HD, Zimmermann EA, Schaible E, Tang SY, Alliston T, Ritchie RO.
Characterization of the effects of x-ray irradiation on the hierarchical struc-
ture and mechanical properties of human cortical bone. Biomaterials
2011;32(34):8892e904.
[40] Bramanti E, Ferrari C, Angeli V, Onor M, Synovec RE. Characterization of BSA
unfolding and aggregation using a single-capillary viscometer and dynamic
surface tension detector. Talanta 2011;85(5):2553e61.
2004. p. 427e541.
[
[
3] Poon CS, Huang Q, Fung PC. Degradation kinetics of cuprophenyl yellow RL by
UV/H2O2/ultrasonication (US) process in aqueous solution. Chemosphere
1999;38(5):1005e14.
4] Bartalucci G, Bianchini R, Catelani G, D'Andrea F, Guazzelli L. Naturalised dyes:
a simple straightforward synthetic route to a new class of dyes e glyco-
azodyes (GADs). Eur J Org Chem 2007:588e95.
[
[
[
5] Bianchini R, Corsi M, Bonanni M. Italian patent FI2013A000093, Coloranti
naturalizzati con l'unit aꢀ deossi (piperazinil) lattosio.
b
ꢀGlicosídeoUm ꢀGlicosídeo 23. Atti R Accad Naz Lincei
a
6] Corsi M, Bonanni M. “Colbiotech Project”, POR CREO FSE 2007-2013 Ob. 2
fellowship. The University of Florence; 2011.
7] Monti S, Bramanti E, Della Porta V, Onor M, D'Ulivo A, Barone V. Interaction of
collagen with chlorosulphonated paraffin tanning agents: fourier transform
infrared spectroscopic analysis and molecular dynamics simulations. Phys
Chem Chem Phys 2013;15(35):14736e47.
[
8] Nashy E-SHA, Hussein AI, Essa MM. Retanning agents for chrome tanned
leather based on emulsion nano-particles of styrene/butyl acrilate co-
polymers. N Y Sci Int 2010;3(11):13e21.
9] Gautieri A, Vesentini S, Redaelli A, Buehler MJ. Viscoelastic properties of model
segments of collagen molecules. Matrix Biol 2012;31:141e9.
[41] Bryan MA, Brauner JW, Anderle G, Flach CR, Brodsky B, Mendelsohn R. FTIR
studies of collagen model peptides: complementary experimental and
[
simulation approaches to conformation and unfolding.
J Am Chem Soc
[
[
[
10] Chang SW, Shefelbine SJ, Buehler MJ. Structural and mechanical differences
between collagen homo and heterotrimers: relevance for the molecular origin
of brittle bone disease. Biophys J 2012;102:640e8.
11] Uzel SGM, Buehler MJ. Molecular structure, mechanical behavior and failure
mechanism of the C-terminal cross-link domain in type I collagen. J Mech
Behav Biomed Mater 2011;4:153e61.
12] Russo A, Gautieri A, Vesentini S, Redaelli A, Buehler MJ. Coarse-grained model
of collagen molecules using an extended MARTINI force field. J Chem Theory
Comput 2010;6:1210e8.
2007;129(25):7877e84.
[42] Sellaro TL, Hildebrand D, Lu Q, Vyavahare N, Scott M, Sacks MS. Effects of
collagen fiber orientation on the response of biologically derived soft tissue
biomaterials to cyclic loading. J Biomed Mater Res Part A 2007;80(1):
194e205.
[43] Berto M. Ausiliari di tintura. Appunti di chimica conciaria. 2004.
[44] Mitra T, Sailakshmi G, Gnanamani A, Mandal AB. Studies on cross-linking of
succinic acid with chitosan/collagen. Mater Res 2013;16:755e65.
[45] Farlay D, Duclos M-E, Gineyts E, Bertholon C, Viguet-Carrin S, Nallala J, et al.
ꢀ
1
[
13] Smith JW. Molecular pattern in native collagen. Nature 1968;219:157e8.
14] Orgel JP, Irving TC, Miller A, Wess TJ. Microfibrillar structure of type I collagen
in situ. Proc Natl Acad Sci U S A 2006;103:9001e5.
The ratio 1660/1690 cm measured by infrared microspectroscopy is not
specific of enzymatic collagen cross-links in bone tissue. PLoS ONE
2011;6(12). e28736.
[
[
[
[
[
[
15] Smith BC. Fundamentals of fourier transform infrared spectroscopy. Boca
Raton, Florida. 1996.
16] Elliott A, Ambrose EJ. Structure of synthetic polypeptides. Nature
[46] Bramanti E, Lenci F, Sgarbossa A. Effects of hypericin on the structure and
aggregation properties of b-amyloid peptides. Eur Biophysics J 2010;39(11):
1493e501.
1950;165(4206):921e2.
[47] Hunter CA, Sanders JKM. The nature of.pi.e.pi. interactions. J Am Chem Soc
1990;112(14):5525e34.
[48] Project E. LIFEþ ENV/IT/000352 “Naturalised dyes replacing commercial col-
orants for environmentally friendly leather dyeing and water recycle-Bio-
NaD”. 2014. www.lifebionad.com.
17] Krimm S, Bandekar J. Vibrational spectroscopy and conformation of peptides,
polypeptides, and proteins. Adv Protein Chem 1986;38:181e364.
18] Susi H, Byler DM. Resolution-enhanced fourier transform infrared spectros-
copy of enzymes. Methods Enzymol 1986;130:290e311.
19] Surewicz WK, Mantsch HH. New insight into protein secondary structure
from resolution-enhanced infrared spectra. Biochim Biophys Acta 1988;952:
[49] Allegre C, Moulin P, Maisseu M, Charbit F. Savings and re-use of salts and
water present in dye house effluents. Desalination 2004;162:13.
115e30.