Content uploaded by Koen Peter Vercruysse
Author content
All content in this area was uploaded by Koen Peter Vercruysse on Jan 21, 2025
Content may be subject to copyright.
1
Disorder vs. order: the common features in the spectroscopic analyses of
diverse melanin materials
Koen P. Vercruysse; Chemistry Department; Tennessee State University
kvercruysse@tnstate.edu
Abstract
Melanin materials were synthesized using a variety of precursors. Crude reaction mixtures and
select purified materials were evaluated using absorbance spectroscopy from the UV to the
visible and near-infrared regions of the electromagnetic spectrum. Combining new and previous
experimental results, 968 different reactions were evaluated objectively for their type of
appearance as eumelanin- or pheomelanin-like. In addition, two mathematical models were
compared to evaluate the intensity of the color associated with the melanin material. One based
upon the commonly used exponential equation, and another based on a reciprocal power
equation. The results obtained from both approaches yield similar results: objective, numerical
parameters, combining concentration- and material-specific factors, to compare different melanin
materials built from varying types of precursors. FT-IR spectroscopic analyses of select purified
materials highlight the common features of the melanin materials, independent of the precursor
employed. In addition, FT-IR spectroscopy was used to evaluate the presence of carboxylic acid
or carboxylate functional groups. Such features may be present due to the precursor employed
and/or due to the oxidative cleavage of aromatic rings that may occur during the synthesis of the
material. Depending on the purification process involved, the melanin material may contain
carboxylates (dialysis) or carboxylic acids (HCl-induced precipitation or acidification followed
by dialysis).
Keywords
eumelanin; pheomelanin; absorbance spectroscopy; FT-IR spectroscopy; catecholamines;
catechols; serotonin
1. Introduction
Melanin (MN) is a term used to describe a diverse group of biomolecules present in many
groups of living organisms and has been the subject of various reviews. Despite decades of
research, MN is still poorly defined in its chemical structure and therefore the review reports on
MN offer diverse descriptions of what constitutes MN.[1-7] In human and other animal species’
physiology two distinct classes of MN are responsible for the coloration of skin and hair:
eumelanin (EuMN) and pheomelanin (PhMN).[8,9] EuMN is typically described as a brown to
black colored material built from L-DOPA as the precursor. PhMN is commonly described as a
yellow to red pigment formed by a combination of L-DOPA and the amino acid cysteine. This
distinction between EuMN and PhMN may only be applicable for in vivo biosynthesis. In vitro,
2
the addition of L-cysteine to the air-mediated oxidation of MN precursors in alkaline
environment leads to darker, more EuMN-like materials.[10] On the other hand, yellow-to-red
materials can be generated in vitro if other precursors, e.g., epinephrine, are used, or may be
hiding behind the dark-colored, EuMN-like material.[10,11] Apart from L-DOPA, other
catecholamines like dopamine or norepinephrine, with or without cysteine, have been described
as the precursors of neuromelanin, a MN material found in the brain of humans or animals.[12]
Plant, fungal and some bacterial MN are built from nitrogen-free precursors, e.g., homogentisic
acid, catechol or dihydroxynaphthalene.[2,3,13,14] The experiments described in this report
involve MN-like materials synthesized from a wide variety of catecholic (ortho diphenolic)
precursors. An exception to this is the use of serotonin in select experiments. Serotonin-related
experiments are included in this report for a) their unique results, b) the physiological importance
of this compound, c) the similarities in the final product generated compared to the other
precursors, and d) the precedent set in other reports.[15,16] In addition, the term “MN-like” used
in this report is solely based upon the appearance of any material as dark-colored, EuMN-like, or
as light-colored, PhMN-like, as has been done in previous reports.[10,17]
A striking feature of MN materials is their capacity to absorb light in a monotonic,
exponentially declining fashion through the ultraviolet (UV), visible (vis) and near infrared
(NIR) regions of the electromagnetic spectrum. This phenomenon creates their appearance as
dark-colored, EuMN-like, or light-colored, PhMN-like, and anything in between.[10,17] This
monotonic absorbance spectrum has been explained in terms of an overlay of the absorbance
spectra of individual chemical species present in the material; the so-called chemical diversity
model.[18-20] An alternative model invokes a combination of geometric order and disorder within
MN aggregated structures, leading to delocalized excitonic effects amongst the oligomeric units
making up the aggregates.[21] However, others have argued that the monotonic absorbance
spectra observed in MN materials are due to charge transfer (CT) phenomena.[22,23] An excellent
review and discussion on the interaction of MN materials with electromagnetic radiation is given
by Xie et al.[6]
Despite the fact that virtually all review papers will highlight MN’s insolubility in water and
organic solvents, MN materials are often described as “soluble” in alkaline, e.g., KOH or NaOH,
environments.[3,24] The synthesis process used in our experiments involves the use of Na2CO3 as
the base and the MN products generated almost always present themselves as seemingly
3
dissolved in the medium. It has been argued that the MN materials thus generated are dispersions
of a colloidal, hybrid material.[25-27] A model has been proposed that in vitro synthesized MN
consists of a dark-colored core MN-like material physically stabilized by a colorless ligand. This
colorless ligand has been observed to strongly absorb UV light and to have fluorescent
properties.[26] The physical stability or instability of a colloidal particle depends on physical and
chemical factors (particle size, pH, ionic strength, surface charge, zeta potential).[28] Thus,
depending on the chemical nature of the precursor, the addition of HCl or the addition of mono-
or multivalent cations, MN materials may precipitate or remain dispersed in the medium.[25-27]
The notion that MN materials exist as colloidal particles is nothing new and many applications
are explored on the basis of the colloidal properties of MN materials.[3,6,14,29-37] In this context
it is worth noting that, despite the uncertainties regarding the chemical structure(s) behind MN
materials, there seems to be a consensus that MN materials are supramolecular structures
generated through a multi-step process, proceeding from oligomers to aggregation into proto-
particles, and the final, so-called, type A and type B particles.[38,39]
Most studies related to MN-like materials focus on the use of dopamine or L-DOPA, or their
associated derivatives 5,6-dihydroxyindole (DHI) or 5,6-dihydroxyindole carboxylic acid
(DHICA), as precursor for the synthesis of MN. The approach in this and previous
reports[10,17,26,40] is different by using a broad variety of precursors to synthesize MN-like
materials (see Figure 1) and compare some of their properties.
Figure 1. Chemical structures of the precursors used in the synthesis of MN-like materials.
4
MN-like materials were generated through air-mediated oxidation in an alkaline environment
and were evaluated for their capacity to absorb light from the UV to the NIR region of the
electromagnetic spectrum. In addition, select materials were studied using FT-IR spectroscopy
following dialysis or HCl-induced precipitation or acidification. The use of different precursors
to generate MN-like materials allowed us to highlight the features that MN-like materials have in
common. The results presented in this report are a combination of recently conducted, unreported
experiments, and an evaluation of data accumulated from experiments reported upon between
2018 and 2024.[10,11,17,25-27,40-43] This report highlights two important features. First, the use of
two mathematical equations, and their associated parameters, to propose a process for an
objective graphical or numerical description and comparison of the appearance of any MN-like
material. Second, the use of FT-IR spectroscopic analyses to evaluate the presence of carboxylic
acid or carboxylate functional groups. Carboxylic acids/carboxylates may be present as part of
the precursor employed (see Figure 1), and/or due to the oxidative cleavage of aromatic rings
that may occur during the synthesis of the material.[44] In addition, we demonstrate that the
purification process employed will determine whether the carboxylic acid functional groups are
present in their acid or conjugate base form.
2. Materials and Methods
2.1 Materials
Dopamine.HCl (Thermo Scientific. Waltham, MA, USA), L-DOPA (MP Biomedicals, Santa
Ana, CA, USA), epinephrine (Alfa Aesar, Ward Hill, MA, USA), norepinephrine.HCl (Sigma, St
Louis, MO, USA), catechol (Acros Organics, Geel, Belgium), pyrogallol (Sigma, St Louis, MO,
USA), caffeic acid (Sigma, St Louis, MO, USA), chlorogenic acid (Alfa Aesar, Waltham, MA,
USA), 3,4-dihydroxybenzoic acid (Aldrich, St Louis, MO, USA), serotonin.HCl (Thermo
Scientific, Waltham, MA, USA), and 5,6 dihydroxyindole (DHI; Thermo Scientific, Waltham,
MA, USA) were all purchased through Fisher Scientific (Waltham, MA, USA) but originated
from different suppliers as indicated.
2.2 UV, vis/NIR, UV and fluorescence spectroscopy
UV, vis/NIR and fluorescence spectroscopic measurements were made in wells of a 96-well
microplate using a SynergyHT microplate reader from Biotek (Winooski, VT, USA). For
5
measurements involving absorbance readings below 350 nm, UV-transparent microplates were
used. Spectra were recorded with 5 nm intervals. The spectrophotometer used in our experiments
provides reproducible and linear, concentration-dependent measurements for absorbance
readings up to 3.5. Fluorescence measurements were made in wells of an opaque 96-well
microplate with excitation filter set at 360 nm, emission filter set at 460 nm, and sensitivity
factor set at 75.
2.3 FT-IR spectroscopy
FT-IR analyses were performed using the Spectrum Two FT-IR spectrometer from
PerkinElmer (Waltham, Massachusetts). Scans were made using the universal ATR accessory
between 650 and 4,000 cm-1 with a resolution of 4 cm-1 and using the OptKBr beam splitter and
LiTaO3 detector. For each sample 24 scans were accumulated.
2.4 Evaluation of crude reaction mixtures
Over the course of seven years (2018-2024), reactions involving the air-mediated oxidation of
catecholic precursors in the presence of Na2CO3 were studied for varying reasons.[10,11,17,25-
27,40-43] The reaction conditions differed in: a) the precursor used (see Figure 1), b) the
concentration of precursor used (between 0.5 and 20 mM), c) the concentration of Na2CO3
present (between 5 and 50 mM, d) whether the mixture was kept stagnant or stirring, e) the
reaction volume (between 200 L and 200 mL) or f) temperature (room temperature or 37˚C).
RP-HPLC was used to monitor the disappearance of precursor as detailed elsewhere.[26] At the
end of the reaction an absorbance scan of the crude reaction mixture in the vis/NIR range (350
nm to 900 nm) was recorded.
2.5 Evaluation of purified materials
Reactions were set up by dissolving between 200 and 300 mg precursor in 100 mL 25 or 50
mM Na2CO3. A higher concentration of Na2CO3 was used for the precursors in hydrochloride
salt form. A reaction with epinephrine was set up by dissolving 250 mg in 50 mL 5 mM HCl
followed by the addition of 50 mL 200 mM Na2CO3. A reaction involving serotonin was set up
by dissolving 100 mg serotonin.HCl in 40 mL Na2CO3 at 50 mM. This mixture was kept stagnant
at room temperature for up to ten days. The reaction was monitored visually, and photographs
were taken on various occasions.
Crude reaction mixtures were purified through dialysis of the reaction mixture followed by
freeze drying. Dried materials were scanned using FT-IR spectroscopy. Select dried materials
6
were dispersed in distilled water at a concentration between 1 and 2 mg/mL and a dilution series
was prepared in distilled water. The diluted mixtures were scanned for their absorbance in the
vis/NIR and UV range. Select dialyzed and dried materials were dispersed in water at a
concentration of 6 mg/mL. This dispersion was acidified using HCl to a final concentration of
0.1N. The precipitations were centrifuged, washed with water and freeze-dried. When no or little
precipitation occurred, the acidified dispersion was dialyzed and freeze-dried. FT-IR spectra
were recorded of all dried, HCl-induced precipitated or acidified materials.
2.6 Dialysis and freeze drying
Dialysis was performed using Spectrum Spectra/Por RC dialysis membranes with molecular-
weight-cut-off (MWCO) of 3.5kDa obtained from Fisher Scientific (Suwanee, GA). Samples
were freeze-dried using a Labconco FreeZone Plus 4.5L benchtop freeze dry system obtained
from Fisher Scientific (Suwanee, GA).
3. Results
3.1 Observations on reaction with serotonin
Reactions involving serotonin were unique amongst all precursors studied thus far. Color
formation and precursor disappearance, as judged from RP-HPLC analyses, proceeded much
slower compared to the other precursors tested. MN formation never yielded a physically stable
dispersion but always resulted in extensive precipitations and coating of the reaction containers
as shown elsewhere.[26] When kept under stagnant conditions a dark film formed on the top of
the reaction mixture when viewed perpendicularly. However, when viewed from a different angle
iridescent colors could be observed. The color of this film appeared to change as the reaction
proceeded. These observations are illustrated in Figure 2, panels A through D.
A B C D
Figure 2. Photographs of the top view of a stagnant reaction mixture containing serotonin (see Section 2.5). A)
Perpendicular view after 24 hours of reaction. B) Angled view after 24 hours of reaction. C) Angled view after 26
hours of reaction. D) Angled view after 7 days of reaction. (Photographs by K.V.)
7
After 7 days the mixture was stirred, and significant amounts of precipitation were observed.
The mixture was centrifuged, and the colorless supernatant was kept at room temperature for an
additional 7 days. Color and film formation appeared anew. However, much less iridescent colors
were observed in this secondary film. These results suggest that the film formation that occurs
when the reaction mixture is kept stagnant may block or slow down the infusion of air (and
oxygen) into the reaction mixture. This phenomenon could account for the slower and
incomplete progression of the reaction. Film formation was also observed for similar stagnant
reactions involving DHI or dopamine, but these did not appear to show any significant iridescent
colors and did not prevent the reaction from proceeding to completion. For all other precursors
dark colors appeared at the interface of the air/reaction mixture when similar reactions were kept
stagnant. However, the dark-colored materials slowly diffused into the lower parts of the reaction
mixture and ultimately formed a homogenous-looking mixture.
3.2 Mathematical analyses of absorbance profiles
As documented in the literature, the absorbance spectra of MN materials dispersed in water
yield an exponentially declining profile with increasing wavelength; particularly in the vis/NIR
region of the electromagnetic spectrum. Thus, such absorbance spectra can be fitted with an
exponential function as shown in Eq.1.
Eq.1
In this equation, k is the decay constant of the exponential profile of the absorbance (A) as a
function of the wavelength (). A0 is the absorbance of the sample when = 0. Exponential
regression analyses were typically performed for absorbance readings between 500 and 800 nm
to avoid the interference of components with strong absorbance in the upper UV region (300 –
400 nm) and to ensure that all absorbance readings were sufficiently above baseline. For
particularly light-colored materials, e.g., derived from epinephrine or pyrogallol, regression
analyses were performed between 500 and 700 nm. Alternatively, the absorbance profiles could
be modeled with a reciprocal power function as shown in Eq.2.
+ b Eq.2
8
Empirically we observed that the value of z in Eq.2, which gives the best fit through the
experimental data, is related to the value of k from Eq.1. Thus, for any value of k, there exists a
value of z for which the relationship shown in Eq.3 holds.
Eq.3
The relationship between k from Eq.1 and z from Eq.2 was studied empirically. By arbitrarily
fixing the value of A0 from Eq.1 at 10 (after it had been determined that the process described
below is independent of the value of A0), and varying the value of k between 0.002 and 0.02 with
0.0025 intervals, the value of z that best fits Eq. 3 was sought for any value of k. Thus, for any
set of A0 and k, A was calculated as a function of λ (between 500 and 800 nm) according to Eq.
1. A was plotted as a function of 1/λz according to Eq.2 for values of z between 0.05 and 10 with
0.05 intervals. The best value of z was chosen based upon the value of the correlation coefficient
(r2) of the linear regression analysis according to Eq.2. Figure 3, panel A, shows plots of r2 as a
function of z for k = 0.0065 or 0.013. Thus, for k = 0.0065, the best value of z is 3.00 and for k =
0.0130, the best value of z is 6.80. As mentioned above, this process was repeated for values of k
between 0.002 and 0.02, and the optimal values of z were plotted as a function of the
corresponding values of k as shown in Figure 3, panel B.
A B
Figure 3. A) Values of r2 as a function of z from Eq.2 of the linear regression analysis of the values of A as function
of 1/z for two values of k from Eq.1 with A0 set at 10. B) Empirical relationship between the value of k from Eq.1
and the value of z from Eq.2 that best fits the relationship shown in Eq.3.
As shown in Figure 3, panel B, an apparent linear relationship exists between the values of k and
z. With this relationship between k and z, the value of z can be estimated from the value of k that
9
best fits the experimental data according to Eq.1. The application of these mathematical
processes is illustrated with two examples in Figure 4, panels A and B. The results in panel A are
from a mixture containing L-DOPA-based MN, while the results in panel B are from a mixture
containing norepinephrine-based MN.
A B
Figure 4. Comparison of the modelling of experimental absorbance readings obtained from, A) an L-DOPA-based
MN material, or B) a norepinephrine-based MN material, with an exponential function (Eq.1; black) or reciprocal
power function (Eq.2; red).
To objectively quantify the intensity of the color of any given MN material, Eq.1 and Eq.2 are
integrated between 400 and 900 nm using the established values of A0 and k, or a, b and z
respectively. Such integrations, shown in Eq.4 and 5, represent the area-under-the-curve, termed
AUCvis, and represent the total absorbance of the spectrum within the vis/NIR region of the
electromagnetic spectrum.
Eq.4
Eq.5
The value of A0 in Eq. 1 depends on the concentration of the sample, while the value of k is
independent of the concentration and is to be considered a material specific parameter. Similarly,
the values of a and b in Eq.2 depend on the concentration of the sample, while the value of z is
independent of the concentration and is to be considered a material specific parameter. These
facts are illustrated in Figure 5, panels A through D, for a dialyzed and freeze-dried, L-DOPA
based MN material. Figure 5, panel A, shows the absorbance profiles in the vis/NIR region of the
same sample dispersed at varying concentrations. Figure 5, panel B, compares the values of
10
AUCvis, calculated using Eq.4 or Eq.5, as a function of sample concentration. Figure 5, panel C,
shows the linear relationship between the value of A0 from Eq.1 as a function of sample
concentration and includes the average value (± standard deviation; n = 7) of k from Eq.1. Figure
5, panel D, shows the linear relationship between the values of a and b from Eq.2 as a function of
sample concentration and includes the average value (± standard deviation; n = 7) of z from
Eq.2.
A B
C D
Figure 5. A) Absorbance profiles in vis/NIR region of L-DOPA based MN material dispersed in water at varying
concentrations. B) Comparison of the values of AUCvis calculated using Eq.4 or Eq.5 as a function of sample
concentration. C) Linear relationship between the value of A0 from Eq.1 as a function of sample concentration and
average value (± standard deviation; n = 7) of k from Eq.1. D) Linear relationships between the values of a (red) or b
(blue) from Eq.2 as a function of sample concentration and average value (± standard deviation; n = 7) of z from
Eq.2.
11
The absorbance in the UV region by MN materials can not readily be approached with a simple
mathematical equation due to the presence of specific absorbance bands within that region. To
obtain an objective estimate of the total absorption in the UV region of the electromagnetic
spectrum (between 230 and 400 nm), the AUC in that region, termed AUCUV, was calculated as
shown in Eq.6.
Eq.6
Figure 6, panel A, shows the absorbance profiles in the UV region of the same samples shown
in Figure 5. Figure 6, panel B, shows the linear relationship between the value of AUCUV and the
concentration of the sample.
A B
Figure 6. A) UV absorbance profiles of a L-DOPA based MN material dispersed in water at varying concentrations.
B) Relationship between the values of AUCUV calculated using Eq.6 and sample concentration.
A parameter like AUCvis provides an objective evaluation of the intensity of the appearance
(color or “darkness”) of a particular MN dispersion. However, it does not provide an objective
evaluation of the type of MN material: EuMN- or PhMN-like. To objectively evaluate the nature
of the MN dispersion, the ratio of the absorbance at 650 nm over the absorbance at 500 nm
(A650/A500) was evaluated as a function of the value of k from Eq.1 as has been discussed
elsewhere.[10] Although these two wavelengths are arbitrarily chosen, it follows precedent as a
suggested way to differentiate dark-colored EuMN from light-colored PhMN.[45-47] Given the
exponential relationship between A and (see Eq.1), an exponential relationship exists between
k and A650/A500 as shown in Eq.7.
12
Eq.7
Higher values of k and lower values of A650/A500 typically indicate light-colored (orange to
yellow) materials. Lower values of k and higher values of A650/A500 typically indicate dark-
colored (brown to black) materials. As discussed earlier, over the course of about seven years
(2018-2024), reactions involving the air-mediated oxidation of a variety of precursors were
studied (n = 968). At the end of each reaction an absorbance scan between 350 and 900 nm was
recorded and modeled according to Eq.1. Reactions involving serotonin consistently resulted in
precipitated materials. However, after washing with water the dark-colored material could be
dispersed in 1N HCl. Absorbance spectra could then be recorded and modeled according to Eq.1.
Figure 7 shows a plot of the average value of k (± standard deviation) vs. the average value of
A650/A500 (± standard deviation) for every individual precursor studied. These data points are
plotted in relationship to the theoretical line according to Eq.7.
Figure 7. Plot of the average (± standard deviation) value of k as a function of the average (± standard deviation)
value of A650/A500 according to Eq.7 at the end of the synthesis reaction for the individual precursors used in the
studies outlined in Section 2.4.
13
All experimental data points fall on or are close to the theoretical line. An exception appeared to
be some reactions involving catechol for which the average values of k and A650/A500 are
sometimes off the theoretical line. The absorbance profiles of reactions involving catechol
contain the typical monotonic profiles of MN materials but with a weak, broad absorbance band
around 650 nm (results not shown), and this accounts for some of the deviations observed in
Figure 7. Figure 8 presents a plot of all individual data points (n = 968) associated with the
average values shown in Figure 7. In this figure the data points associated with the experiments
involving catechol were separated from the others.
Figure 8. Plot of the individual values of k as a function of the individual values of A650/A500 according to Eq.7
associated with the results presented in Figure 7. The results associated with the experiments involving catechol
were separated from the other results.
Figure 8 confirms the pattern of results shown in Figure 7, particularly the deviation of some
results obtained with catechol from the theoretical line associated with Eq.7. In addition, the
results shown in Figure 8 suggest the presence of two data clusters as indicated with arrows in
Figure 8. Figure 9 presents a histogram profile of all the values of k shown in Figure 8.
14
Figure 9. Histogram plot of all the individual values of k shown in Figure 8.
The pattern of results shown in Figure 9 suggests a bimodal distribution of the values of k. This
correlates with the visual appearance of the materials as either dark-colored (low value of k) or
light-colored (high value of k) materials, and correlates with a subjective classification, based
upon their appearances, as either EuMN-like (brown to black) or PhMN-like (dark orange to
yellow).
3.3 Characterization of purified materials: vis/NIR and UV absorbance
The color or appearance of MN materials depends on the concentration (C) of the dispersion
and the type of precursor used in its synthesis. This is illustrated in Table 1, showing photographs
of dialyzed and freeze-dried MN materials, built from different precursors, dispersed in water at
different concentrations.
15
Table 1. Photographs of dialyzed and freeze-dried MN materials synthesized from different precursors and
dispersed in water at different concentrations (C); photographs by K.V.)
Pyrogallol-based
C = 472 g/mL
C = 810 g/mL
C = 1,417 g/mL
C = 2,125 g/mL
C = 2,833 g/mL
Dopamine-based
C = 132 g/mL
C = 362 g/mL
C = 925 g/mL
C = 1,388 g/mL
C = 2,775 g/mL
DHI-based
C = 46 g/mL
C = 90 g/mL
C = 175 g/mL
C = 560 g/mL
C = 2,800 g/mL
The photographs shown in Table 1 illustrate the impact concentration and type of MN has on the
appearance of the MN dispersion. Any comparison of the appearance of MN materials should
consider the concentration of MN material involved. Thus, a parameter like AUCvis/C is
proposed as an objective numerical value reflecting the intensity of the color of a particular MN
material. Similarly, a parameter like AUCUV/C is proposed as an objective evaluation of the
capacity to absorb UV light by any MN material. A comparison of such parameters is shown in
Table 2 for dialyzed and freeze-dried MN materials synthesized from a variety of precursors.
16
Table 2. Comparison of AUCvis/C and AUCUV/C values of MN-like materials synthesized from various
precursors. AUCvis was calculated according to Eq.4 and AUCUV was calculated according to Eq.6. Dilution
series of each MN material, dispersed in distilled water, at six (AUCvis) or four (AUCUV) different concentrations
were analyzed in triplicate. All values shown are average ± standard deviation. Some precursors were used twice
to generate a separate batch of MN-like material.
Precursor
AUCvis/C (n = 18)
AUCUV/C (n = 12)
3,4-dihydroxy benzoic acid
catechol
catechol
DHI
dopamine
dopamine
epinephrine
L-DOPA
L-DOPA
norepinephrine
norepinephrine
pyrogallol
pyrogallol
0.090 ± 0.016
0.37 ± 0.03
0.21 ± 0.02
2.23 ± 0.3
0.52 ± 0.05
0.46 ± 0.04
0.022 ± 0.001
0.88 ± 0.05
1.04 ± 0.10
0.35 ± 0.03
0.60 ± 0.01
0.30 ± 0.02
0.29 ± 0.01
1.05 ± 0.14
1.50 ± 0.11
1.10 ± 0.06
2.67 ± 0.33
1.48 ± 0.14
1.62 ± 0.16
1.46 ± 0.05
1.72 ± 0.02
2.17 ± 0.18
1.48 ± 0.05
1.70 ± 0.04
1.32 ± 0.06
1.52 ± 0.03
Other parameters within or associated with Eq.1 or 2 are independent of the concentration of the
sample and can be considered material specific. Such parameters include values of k (Eq.1) or z
(Eq.2). The values of these parameters are associated with the appearance of the MN-like
materials as EuMN- or PhMn-like. Lower values of k or z are associated with EuMN-like
appearance. Higher values of k or z are associated with PhMN-like appearance. Tabel 3 shows
the values of these parameters for the materials listed in Table 2.
17
Table 3. Comparison of k (Eq.1) and z (Eq.2) values of MN-like materials synthesized from various precursors.
All values shown are average ± standard deviation (n = 18) and were derived from the analyses of the materials
listed in Table 2.
Precursor
k
z
3,4-dihydroxy benzoic acid
catechol
catechol
DHI
dopamine
dopamine
epinephrine
L-DOPA
L-DOPA
norepinephrine
norepinephrine
pyrogallol
pyrogallol
0.0127 ± 0.0002
0.0085 ± 0.0004
0.0069 ± 0.0002
0.0050 ± 0.0002
0.0077 ± 0.0005
0.0083 ± 0.0002
0.0128 ± 0.0016
0.0063 ± 0.0002
0.0064 ± 0.0001
0.0118 ± 0.0011
0.0103 ± 0.0001
0.0127 ± 0.0002
0.0130 ± 0.0005
6.54 ± 0.13
4.13 ± 0.23
3.22 ± 0.12
2.14 ± 0.13
3.70 ± 0.26
4.03 ± 0.10
6.58 ± 0.89
2.90 ± 0.10
2.94 ± 0.08
6.02 ± 0.60
5.16 ± 0.03
6.52 ± 0.14
6.72 ± 0.29
In addition to its concentration, the appearance of any MN material depends on its characteristic
as EuMN- or PhMN-like. Thus, the value of (AUCvis/C)/k may be the most suitable parameter
for an objective comparison of MN-like materials. The values of this parameter are shown in
Table 4 for the materials listed in Table 2.
Table 4. Comparison of (AUCvis/C)/k values of MN-like materials synthesized from various precursors. All
values shown are average ± standard deviation (n = 18) and are based upon the analyses of the materials listed in
Tables 2 and 3.
precursor
(AUCvis/C)/k
3,4-dihydroxy benzoic acid
catechol
catechol
DHI
dopamine
dopamine
epinephrine
L-DOPA
L-DOPA
norepinephrine
norepinephrine
pyrogallol
pyrogallol
7.1 ± 1.2
44 ± 5.2
30 ± 3.6
447 ± 76
67 ± 6.2
56 ± 5.0
1.69 ± 0.09
139 ± 9.3
163 ± 14
30 ± 0.88
58 ± 0.68
24 ± 1.40
22 ± 0.67
18
All the parameters listed in Tables 2, 3 and 4 indicate that the MN material generated from DHI
was the darkest in appearance and the MN material generated from epinephrine was the lightest
in appearance. Comparing the values of AUCvis/C (see Table 2), the DHI-based material could be
considered 101X darker than the epinephrine-based material. Alternatively, comparing the values
of (AUCvis/C)/k (see Table 4), the DHI-based material could be considered 265X darker than the
epinephrine-based material. However, comparing the values of AUCUV/C (see Table 2), the DHI-
based material absorbed only 1.83X more UV light compared to the epinephrine-based material.
3.4 Characterization of purified materials: FT-IR spectroscopy
Materials derived through dialysis of the crude reaction mixture were scanned using FT-IR
spectroscopy. The raw data of these scans all exhibited a distinct signal around 1,580 cm-
1indicative of the so-called ring mode of aromatic rings.[48] Thus, all spectra were normalized for
their peak associated with this ring mode. Figure 10 shows an overlay of select FT-IR profiles
obtained from MN materials derived from different precursors following the dialysis of the crude
reaction mixture.
19
Figure 10. Normalized FT-IR spectra of select MN-like materials derived from various precursors. The insert shows
the profile obtained from a catechol-based material. The numbered arrows indicate common signals in all spectra
and are discussed in the text.
The numbered arrows indicate peak maxima observed in all FT-IR profiles, independent of the
precursor involved in the synthesis. The average (± standard deviation; n = 15) peak maxima
were observed at: 1) 3,215 ± 18 cm-1, 2) 1,583 ± 6 cm-1, 3) 1,377 ± 8 cm-1 and 4) 778 ± 12 cm-1.
These signals could be assigned to: 1) O-H, and, if applicable. N-H stretch, 2) ring mode of
aromatic ring, 3) O-H bend and 4) out-of-plane C-H bend of aromatic ring.[48-50] As will be
discussed later in this report, for materials purified through dialysis, carboxylic acid functional
groups may exist in carboxylate form. Thus, the signals associated with the asymmetric and
symmetric C-O stretch of carboxylates (typically present as a pair of signals between 1,540 to
1,650 cm-1 and between 1,360 to 1,450 cm-1) overlap with some of the features discussed
above.[51] None of the profiles shown in Figure 10 show a distinct signal near 1,700 cm-1 which
would indicate the presence of a carbonyl group.[52] Such a carbonyl group could be associated
20
with the presence of a carboxylic acid, aldehyde or ketone functional group. Even materials
made from L-DOPA or 3,4-dihydroxybenzoic acid do not show such a signal in their FT-IR
spectra. However, given the alkaline conditions of the synthesis process, and dialysis as the
purification process, it was assumed that any carboxylic acid functional group present existed in
its conjugate, carboxylate state. To evaluate this issue, dialyzed materials were dispersed in water
and 1N HCl was added to induce precipitation of the material. Precipitates were washed with
water, freeze-dried and scanned using FT-IR spectroscopy. In the event the material did not
precipitate in the presence of HCl, the acidified mixture was dialyzed, freeze-dried and scanned
using FT-IR spectroscopy. Figure 11, panels A through H, show comparisons of the normalized
FT-IR spectra of the dialyzed and the HCl-precipitated or acidified materials. In addition, the
normalized signal of the dialyzed material was subtracted from the HCl-precipitated or acidified
material, and these profiles are included in Figure 11.
A B
C D
21
E F
G H
Figure 11. A) Normalized FT-IR spectra of dialyzed and HCl-precipitated materials derived from L-DOPA. B)
Absorbance difference following subtraction of normalized spectrum of the dialyzed material from the normalized
spectrum of the HCl-precipitated material shown in panel A. C) Normalized FT-IR spectra of dialyzed and HCl-
acidified materials derived from 3,4-dihydroxybenzoic acid. D) Absorbance difference following subtraction of
normalized spectrum of the dialyzed material from the normalized spectrum of the HCl-acidified material shown in
panel C. E) Normalized FT-IR spectra of dialyzed and HCl-precipitated materials derived from dopamine. F)
Absorbance difference following subtraction of normalized spectrum of the dialyzed material from the normalized
spectrum of the HCl-precipitated material shown in panel E. G) Normalized FT-IR spectra of dialyzed and HCl-
acidified materials derived from catechol. H) Absorbance difference following subtraction of normalized spectrum
of the dialyzed material from the normalized spectrum of the HCl-acidified material shown in panel G.
All the profiles obtained following the HCl-induced precipitation or acidification show a clear
peak or shoulder in the low 1,700 cm-1 region of the spectrum. By comparing the FT-IR spectra
of the dialyzed materials to the HCl precipitated or acidified materials, one can assign the signal
in the low 1,700 cm-1 region to the presence of carboxylic acid groups. The subtracted
absorbance profiles shown in Figure 11, panels, B, D, F and H, not only show the emergence of
22
the signal in the low 1,700 cm-1 region, but additional common features. Figure 12 presents an
overlay of all (n = 7) such subtracted absorbance profiles obtained thus far, including the profiles
in Figure 11, panels, B, D, F and H.
Figure 12. Absorbance difference of normalized FT-IR spectra of dialyzed materials and HCl-induced precipitation
or acidification derived from various precursors. The normalized spectra of the dialyzed materials were subtracted
from the normalized spectra of the HCl-precipitated or acidified materials. The numbered arrows indicate common
features in all these subtracted, normalized FT-IR absorbance profiles.
As shown with arrows in Figure 12, the subtracted, normalized FT-IR spectra show four
common peaks independent of the type of precursor used in the synthesis of the material. The
average (± standard deviation; n = 7) position of these peaks, positive or negative, are: 1*) 1,715
± 5 cm-1, 2*) 1,546 ± 9 cm-1, 3*) 1,383 ± 5 cm-1 and 4*) 1,186 ± 17 cm-1. The positive signal at
1,715 cm-1 can be assigned to the C=O stretch of the emerging carboxylic acid functional
groups.[53] The negative signals at 1,546 and 1,383 cm-1 can be assigned to the disappearance of
the asymmetric and symmetric C-O stretch of the carboxylate functional groups.[51] The positive
signal at 1,186 cm-1 can be assigned to the C-O stretch of the emerging carboxylic acid groups;
possibly overlapping with the C-O stretch of phenols.[53,54] Although the addition of HCl would
23
render any amine group into its protonated salt form, the N-H stretching or bending vibrations
may be too weak to be readily distinguishable in the FT-IR spectra of the complex MN
material.[55] In addition to the observations illustrated in Figure 12, it appeared that the signal
associated with the aromatic ring mode shifted from an average of 1,583 ± 6 cm-1 (see above) to
an average value of 1,601 ± 5 cm-1 (n = 7) in the FT-IR spectra of the HCl precipitated or
acidified materials.
Attempts were made to correlate any of the normalized signals of the FT-IR spectra of the
dialyzed materials with any of the parameters associated with the appearance of the dialyzed
materials as discussed earlier and shown in Tables 1 through 3. The best correlations observed
thus far are shown in Figure 13, panels A through C.
A B
C
Figure 13. Correlation between select normalized FT-IR signals obtained from dialyzed materials listed in Tables 2
through 4, and values of: A) k from Table 3, B) values of AUCvis/C from Table 2 and C) values of (AUCvis/C)/k from
Table 4.
24
The results shown in Figure 13, panels A through C. all point to a possible correlation between
the relative strength of the FT-IR signal between 1,130 and 1,140 cm-1 and the appearance of the
material. The stronger this signal, the darker, as evident from the lower k or the higher AUCvis/C
or (AUCvis/C)/k values, the appearance of the material. This signal may be associated with the C-
O stretch of phenols.[54]
4. Discussion
Based upon absorbance and FT-IR spectroscopy, this report highlights common features in
some of the physical and chemical properties of MN-like materials generated from a wide variety
of precursors. Of all the precursors explored, serotonin stands apart from the others as it does not
contain a dihydroxyphenyl moiety in its structure and thus can not be converted into a quinone
type of moiety which is common when using other precursors.[1] Despite this key difference in
chemistry, serotonin can yield a MN-like material when kept in an alkaline environment and
exposed to air. When dispersed in HCl the material has an appearance like L-DOPA- or
dopamine-based MN (see Figure 7).[26] The FT-IR spectrum of serotonin-based MN has the
same key features as all the other MN materials we evaluated. However, serotonin-based MN
materials behave in an opposite way regarding their physical stability: precipitating in an alkaline
environment and dispersing in an acidic environment. MN materials, and in particular dopamine-
based materials (often referred to as polydopamine or PDA) have received much attention for
their ability to coat surfaces.[56,57] In this context, the properties of serotonin-based MN, despite
their excellent coating abilities, appear to be lacking in attention as few reports describe the use
of this precursor.[15,16]
Although aspects of the mathematical processing of the absorbance spectra described in
Section 3.2 are arbitrarily chosen, the entire process provides an objective framework for the
evaluation and comparison of various types of MN materials. Absorbance profiles of MN
materials have been described as exponential in nature[1], but the use of the corresponding decay
constant k (see Eq.1), nor the possibility for its integration (see Eq.4), have not been explored as
an objective way of comparing different MN materials.[10,25] The use of the reciprocal power
function (see Eq.2), and the calculation of the corresponding material specific parameter z
(Eq.2), is novel. The results presented in Figures 4 and 5, panel B, illustrate that both Eq.1 and
25
Eq.2 are suitable to objectively evaluate the vis/NIR absorbance properties of MN materials. In
this context it is worthwhile pointing out that for some interactions between light and material,
e.g., Rayleigh scattering, a mathematical model involving a reciprocal power factor of is used.
In addition, Eq.2 allows for an estimate of at which the absorbance is 0; termed 0. From Eq.2
it can be derived that 0 = (-a/b)(1/z). Although both Eq.1 and Eq.2 can describe the same
experimental data, and we have empirically established a relationship between the key
parameters k and z (see Figure 3, panel B), providing a theoretical, mathematical framework to
prove the equivalence shown in Eq.3 is beyond the scope of this report (and the mathematical
capabilities of this author). However, it is very likely that the relationship shown in Eq.3 may
only be valid within a limited range of , e.g., between 500 and 800 nm, and not for any range of
values of or within a limited range of values of k, e.g., between 0.002 and 0.020.
Despite the chemical or geometric disorder that may exist within MN materials, the results
shown in Figures 7 and 8 illustrate that the profile of the absorbance of vis/NIR light by MN-like
materials is remarkably consistent. Despite the diversity in precursors used, and the varying
conditions in the 968 reaction mixtures evaluated, the absorbance spectra always exhibited the
exponential aspects described in Eq.1 and Eq.8. This suggests that, despite the varying
experimental conditions, a set of materials were generated that nearly always yielded an
exponential absorbance spectrum with hardly ever any deviations. Any theory or model invoked
to describe the absorbance properties of MN materials would have to explain this observation.
Even when “extra complexity” is introduced to the reaction conditions, e.g., through the addition
of amino acids, the spectra of MN materials obtained in the presence of AA exhibit an
exponential form.[10,17,43] Notable exceptions to this are reactions involving chlorogenic or
caffeic acid and amino acids[41,58,59], or reactions involving catechol and proline or
hydroxyproline.[60] In these cases the materials generated exhibit distinct colors with distinct
absorbance bands in their absorbance spectra. This author has suggested a model that the in vitro
synthesized MN materials are hybrid entities built from at least two different types of
components.[25] Thus, the overall absorbance spectrum of MN materials could be the result of a
synergistic interaction between these two different components. The suggested model echoes
some of the observations made by Mavridi-Printezi et al. regarding the dualistic nature of
PDA.[61] However, in their work both components exhibit broad-band absorbance while in the
author’s model only one component exhibits that spectroscopic feature. More minimalistic, a
26
report by Mavridi-Printezi et al. highlighted that, in the case of dopamine, a simple charge-
transfer adduct formed by a combination of dopamine and its quinone oxidation product, is
sufficient to produce a broad band absorbance spectrum.[62] Wang et al. reported on the synthesis
of a sterically stabilized indolequinone and its absorbance spectrum.[63] This indolequinone
exhibited specific absorbance bands around 400 and 750 nm. Although some argued that this
feature covers the full absorbance range of a typical EuMN material, and that EuMN may consist
of a poly-indoquinone system[1], the absorbance spectrum of the stabilized indolequinone is still
a far cry from the exponential absorbance profile of MN materials. The absorbance spectrum of
the stabilized indolequinone resembles the absorbance spectrum of the green pigment obtained
when chlorogenic acid is oxidized in the presence of amino acids.[41,58,59] Thus, one could argue
that the stabilized indolequinone prevents the material from developing its “natural” exponential,
broad-band absorbance profile.
The results shown in Figures 7 and 8, and the photographs in Table 1, illustrate that through the
choice of precursor and dispersion concentration, appearances ranging from intense black to light
yellow can be obtained. This can be of importance for the development of cosmetic products
aiming to match different shades of skin or hair color.[30,64,65] Objective numerical parameters
as shown in Table 4 could be used to categorize different types of MN materials. Similarly, if UV
absorbance by the MN material is critical, a parameter like AUCUV/C (see Table 2) could be used
to categorize different MN materials. In addition, it could be worthwhile to link any physical
(color, particle size, zeta potential, electron spin) or chemical property (anti-oxidant activity,
cation binding) of a MN-like material to any of the objective parameters listed in Tables 2
through 4. MN materials are frequently discussed in terms of their protective or harmful
interaction with UV light.[6,45] However, despite the markedly darker appearance of DHI
materials compared to epinephrine materials, no markedly different capacity to absorb UV light
was observed (see Section 3.3). This observation strengthens the notion that the color and the
capacity to absorb UV light may be separate properties of the overall MN material; residing in
the two different entities making up the hybrid, colloidal material.[25,26]
Due to its insolubility in water or organic solvent, chemical characterization of MN materials is
still incomplete. A commonly used technique is FT-IR spectroscopy, although it provides only
limited information regarding the chemical details within the MN materials. An overview of
typical peak signals in FT-IR spectra of natural or synthetic MN is presented by Perna et al.[66]
27
The FT-IR analyses we performed on select materials provide various observations. The spectra
of the dialyzed materials shown in Figure 10 have very similar features despite the diversity in
the chemistry of the precursors involved, The FT-IR spectrum of catechol-based materials
exhibit sharper signals (see insert in Figure 10), but the peak maxima are within the same regions
as the peak maxima of the other materials. By comparing and subtracting the FT-IR spectra of
dialyzed vs. HCl-precipitated or -acidified material, one can observe that many MN materials
contain carboxylic acid/carboxylate functionalities (see Figures 11 and 12). Even MN materials
synthesized from precursors without carboxylic acids, e.g., dopamine or catechol, exhibit
carboxylic acid/carboxylate features. This indicates that oxidative aromatic ring cleavage occurs
during the synthesis reaction as has been discussed elsewhere.[44] In the case of amine-
containing MN materials, the amino functionality could be positively charged creating the
possibility that select MN material may exist as zwitterions. The presence or absence of any
charge on the surface of MN materials can impact their physic-chemical properties, e.g., their
tendency to aggregate into smaller or larger units, or their suitability for particular
applications.[67,68]
Conclusions
This report details the common features in the absorbance spectra and FT-IR spectra of a
variety of melanin-like materials synthesized using different precursors and under varying
reaction conditions. Two mathematical equations are explored to fit the experimental absorbance
spectra and to provide objective, numerical parameters to compare and categorize varying
melanin-like materials. FT-IR spectroscopy was used to highlight the common presence of
carboxylic acid or carboxylate functional groups within the melanin-like materials.
Acknowledgements
The author wishes to thank Venise Govan, Jada Harrison, Tyona Caldwell, Stenesha Fortner,
Jaila Winford, Iesha Brown, Sha’Kendria Summers, Jayla Moore, Aaliyah Flake, Keturah Badie,
Alin Joseph and Tylor Miller for the execution and the collection of data associated with the
experiments described herein. JM was supported by the “Bridge to Doctorate Program at
Tennessee State University” grant from the National Science Foundation (NSF; grant 1810991).
VG, JW and KB were supported by The Tennessee Louis Stokes Alliance for Minority
28
Participation (TLSAMP) supported by the National Science Foundation (NSF – HRD 1826954).
AF and TM were supported by NIH grant 5U54CA163066.
References
1. Motovilov, K.A.; Mostert, A.B. Melanin: Nature's 4th bioorganic polymer. Soft Matter 2024, 20, 5635-5651,
doi:10.1039/d4sm00491d.
2. Glagoleva, A.Y.; Shoeva, O.Y.; Khlestkina, E.K. Melanin Pigment in Plants: Current Knowledge and Future
Perspectives. Front Plant Sci 2020, 11, 770, doi:10.3389/fpls.2020.00770.
3. Singh, S.; Nimse, S.B.; Mathew, D.E.; Dhimmar, A.; Sahastrabudhe, H.; Gajjar, A.; Ghadge, V.A.; Kumar, P.;
Shinde, P.B. Microbial melanin: Recent advances in biosynthesis, extraction, characterization, and
applications. Biotechnology Advances 2021, 53, 107773,
doi:https://doi.org/10.1016/j.biotechadv.2021.107773.
4. Solano, F. Melanins: Skin Pigments and Much More—Types, Structural Models, Biological Functions, and
Formation Routes. New Journal of Science 2014, 2014, 498276,
doi:https://doi.org/10.1155/2014/498276.
5. Mostert, A.B. Melanin, the What, the Why and the How: An Introductory Review for Materials Scientists
Interested in Flexible and Versatile Polymers. Polymers (Basel) 2021, 13, doi:10.3390/polym13101670.
6. Xie, W.; Dhinojwala, A.; Gianneschi, N.C.; Shawkey, M.D. Interactions of Melanin with Electromagnetic
Radiation: From Fundamentals to Applications. Chem Rev 2024, 124, 7165-7213,
doi:10.1021/acs.chemrev.3c00858.
7. Qin, Y.; Xia, Y. Melanin in fungi: advances in structure, biosynthesis, regulation, and metabolic
engineering. Microb Cell Fact 2024, 23, 334, doi:10.1186/s12934-024-02614-8.
8. Simon, J.D.; Peles, D.N. The red and the black. Acc Chem Res 2010, 43, 1452-1460,
doi:10.1021/ar100079y.
9. Wakamatsu, K.; Ito, S. Recent Advances in Characterization of Melanin Pigments in Biological Samples.
Int J Mol Sci 2023, 24, doi:10.3390/ijms24098305.
10. Vercruysse, K.P. The “Unconventional” Eect of Cysteine on the In Vitro Synthesis of Melanin. ACS
Omega 2024, doi:10.1021/acsomega.4c00889.
11. Vercruysse, K.P.; Govan, V. Melanogenesis: A Search for Pheomelanin and Also, What Is Lurking Behind
Those Dark Colors? . 2019, doi:https://doi.org/10.26434/chemrxiv.11418075.v1.
12. Nicolaus, B.J. A critical review of the function of neuromelanin and an attempt to provide a unied theory.
Med Hypotheses 2005, 65, 791-796, doi:10.1016/j.mehy.2005.04.011.
13. Suthar, M.; Dufosse, L.; Singh, S.K. The Enigmatic World of Fungal Melanin: A Comprehensive Review. J
Fungi (Basel) 2023, 9, doi:10.3390/jof9090891.
14. Tang, Q.; Li, Z.; Chen, N.; Luo, X.; Zhao, Q. Natural pigments derived from plants and microorganisms:
classication, biosynthesis, and applications. Plant Biotechnol J 2024, doi:10.1111/pbi.14522.
15. Ishino, K.; Nishitani, S.; Man, Y.; Saito, A.; Sakata, T. Surface Characteristics and Formation of
Polyserotonin Thin Films for Bioelectrical and Biocompatible Interfaces. Langmuir 2022, 38, 8633-8642,
doi:10.1021/acs.langmuir.2c01045.
16. Nakatsuka, N.; Hasani-Sadrabadi, M.M.; Cheung, K.M.; Young, T.D.; Bahlakeh, G.; Moshaverinia, A.;
Weiss, P.S.; Andrews, A.M. Polyserotonin Nanoparticles as Multifunctional Materials for Biomedical
Applications. ACS Nano 2018, 12, 4761-4774, doi:10.1021/acsnano.8b01470.
29
17. Vercruysse, K.P.; Govan, V.; Harrison, J.; Caldwell, T.; Winford, J. Appearing as eumelanin or as
pheomelanin and the push towards eumelanin by cysteine. 2022, doi:https://doi.org/10.26434/chemrxiv-
2022-hcz2m.
18. Meredith, P.; Powell, B.J.; Riesz, J.; Nighswander-Rempel, S.P.; Pederson, M.R.; Moore, E.G. Towards
structure-property-function relationships for eumelanin. Soft Matter 2006, 2, 37-44,
doi:10.1039/b511922g.
19. Tran, M.L.; Powell, B.J.; Meredith, P. Chemical and structural disorder in eumelanins: a possible
explanation for broadband absorbance. Biophysical journal 2006, 90, 743-752,
doi:10.1529/biophysj.105.069096.
20. d'Ischia, M.; Napolitano, A.; Pezzella, A.; Meredith, P.; Sarna, T. Chemical and structural diversity in
eumelanins: unexplored bio-optoelectronic materials. Angewandte Chemie (International ed. in English)
2009, 48, 3914-3921, doi:10.1002/anie.200803786.
21. Chen, C.T.; Chuang, C.; Cao, J.; Ball, V.; Ruch, D.; Buehler, M.J. Excitonic eects from geometric order and
disorder explain broadband optical absorption in eumelanin. Nature communications 2014, 5, 3859,
doi:10.1038/ncomms4859.
22. Petropoulos, V.; Mavridi-Printezi, A.; Menichetti, A.; Mordini, D.; Kabacinski, P.; Gianneschi, N.C.;
Montalti, M.; Maiuri, M.; Cerullo, G. Sub-50 fs Formation of Charge Transfer States Rules the Fate of
Photoexcitations in Eumelanin-Like Materials. J Phys Chem Lett 2024, 15, 3639-3645,
doi:10.1021/acs.jpclett.4c00170.
23. Kohl, F.R.; Grieco, C.; Kohler, B. Ultrafast spectral hole burning reveals the distinct chromophores in
eumelanin and their common photoresponse. Chemical Science 2020, 11, 1248-1259,
doi:10.1039/C9SC04527A.
24. Pralea, I.E.; Moldovan, R.C.; Petrache, A.M.; Ilies, M.; Heghes, S.C.; Ielciu, I.; Nicoara, R.; Moldovan, M.;
Ene, M.; Radu, M.; et al. From Extraction to Advanced Analytical Methods: The Challenges of Melanin
Analysis. Int J Mol Sci 2019, 20, doi:10.3390/ijms20163943.
25. Vercruysse, K.; Govan, V.; Winford, J. “Invisible” ligands stabilize colloidal melanin particles – the case of
L-DOPA. 2024, doi:10.26434/chemrxiv-2024-j1z0n D O I: 10.26434/chemrxiv-2024-j1z0n [opens in a new
tab].
26. Flake, A.; Vercruysse, K. The Invisible Fraction within Melanin Capable of Absorbing UV Light and with
Fluorescent Properties: Is It Lacking Consideration? International Journal of Molecular Sciences 2024,
25, 8490.
27. Galeb, H.A.; Eichhorn, J.; Harley, S.; Robson, A.J.; Martocq, L.; Nicholson, S.J.; Ashton, M.D.;
Abdelmohsen, H.A.M.; Pelit, E.; Baldock, S.J.; et al. Phenolic Polymers as Model Melanins.
Macromolecular Chemistry and Physics 2023, 224, 2300025,
doi:https://doi.org/10.1002/macp.202300025.
28. Stenkamp, V.S.; McGuiggan, P.; Berg, J.C. Restabilization of Electrosterically Stabilized Colloids in High
Salt Media. Langmuir 2001, 17, 637-651, doi:10.1021/la001086c.
29. Eom, T.; Ozlu, B.; Ivanova, L.; Lee, S.; Lee, H.; Krajcovic, J.; Shim, B.S. Multifunctional Natural and
Synthetic Melanin for Bioelectronic Applications: A Review. Biomacromolecules 2024, 25, 5489-5511,
doi:10.1021/acs.biomac.4c00494.
30. Biyashev, D.; Siwicka, Z.E.; Onay, U.V.; Demczuk, M.; Xu, D.; Ernst, M.K.; Evans, S.T.; Nguyen, C.V.; Son,
F.A.; Paul, N.K.; et al. Topical application of synthetic melanin promotes tissue repair. npj Regenerative
Medicine 2023, 8, 61, doi:10.1038/s41536-023-00331-1.
31. Jeon, D.-J.; Paik, S.; Ji, S.; Yeo, J.-S. Melanin-based structural coloration of birds and its biomimetic
applications. Applied Microscopy 2021, 51, 14, doi:10.1186/s42649-021-00063-w.
30
32. Mavridi-Printezi, A.; Menichetti, A.; Guernelli, M.; Montalti, M. The Photophysics and Photochemistry of
Melanin- Like Nanomaterials Depend on Morphology and Structure. Chemistry – A European Journal
2021, 27, 16309-16319, doi:https://doi.org/10.1002/chem.202102479.
33. Kohri, M.; Yanagimoto, K.; Kawamura, A.; Hamada, K.; Imai, Y.; Watanabe, T.; Ono, T.; Taniguchi, T.;
Kishikawa, K. Polydopamine-Based 3D Colloidal Photonic Materials: Structural Color Balls and Fibers
from Melanin-Like Particles with Polydopamine Shell Layers. ACS Applied Materials & Interfaces 2018,
10, 7640-7648, doi:10.1021/acsami.7b03453.
34. Liu, Y.; Ai, K.; Liu, J.; Deng, M.; He, Y.; Lu, L. Dopamine-melanin colloidal nanospheres: an eicient near-
infrared photothermal therapeutic agent for in vivo cancer therapy. Adv Mater 2013, 25, 1353-1359,
doi:10.1002/adma.201204683.
35. Park, J.; Moon, H.; Hong, S. Recent advances in melanin-like nanomaterials in biomedical applications: a
mini review. Biomaterials Research 2019, 23, 24, doi:10.1186/s40824-019-0175-9.
36. Patil, A.; Heil, C.M.; Vanthournout, B.; Bleuel, M.; Singla, S.; Hu, Z.; Gianneschi, N.C.; Shawkey, M.D.;
Sinha, S.K.; Jayaraman, A.; et al. Structural Color Production in Melanin-Based Disordered Colloidal
Nanoparticle Assemblies in Spherical Connement. Advanced Optical Materials 2022, 10, 2102162,
doi:https://doi.org/10.1002/adom.202102162.
37. Szpoganicz, B.; Gidanian, S.; Kong, P.; Farmer, P. Metal binding by melanins: studies of colloidal
dihydroxyindole-melanin, and its complexation by Cu(II) and Zn(II) ions. J Inorg Biochem 2002, 89, 45-53,
doi:10.1016/s0162-0134(01)00406-8.
38. Büngeler, A.; Hämisch, B.; Huber, K.; Bremser, W.; Strube, O.I. Insight into the Final Step of the
Supramolecular Buildup of Eumelanin. Langmuir 2017, 33, 6895-6901,
doi:10.1021/acs.langmuir.7b01634.
39. Büngeler, A.; Hämisch, B.; Strube, O.I. The Supramolecular Buildup of Eumelanin: Structures,
Mechanisms, Controllability. International Journal of Molecular Sciences 2017, 18, 1901.
40. Vercruysse, K.P.; Govan, V. The yellow and the black of synthetic melanins. . 2021,
doi:https://doi.org/10.26434/chemrxiv-2021-h9dks.
41. Vercruysse, K.P. Preliminary Observations on the Interactions Between Chlorogenic Acid and Select
Amino Acids. 2019, doi:https://doi.org/10.26434/chemrxiv.8311481.v1.
42. Vercruysse, K.P. Evaluating the “Darkness” of Melanin Materials. 2020,
doi:https://doi.org/10.26434/chemrxiv.12762179.v1.
43. Vercruysse, K.P.; Govan, V.; Fortner, S. Kinetic Study of the Melanogenesis from Select Catecholamines in
the Presence of L-Cysteine or Other Amino Acids. . 2020,
doi:https://doi.org/10.26434/chemrxiv.12678860.v1.
44. Della Vecchia, N.F.; Avolio, R.; Alfè, M.; Errico, M.E.; Napolitano, A.; d'Ischia, M. Building-Block Diversity
in Polydopamine Underpins a Multifunctional Eumelanin-Type Platform Tunable Through a Quinone
Control Point. Advanced Functional Materials 2013, 23, 1331-1340,
doi:https://doi.org/10.1002/adfm.201202127.
45. Ito, S.; Wakamatsu, K.; Sarna, T. Photodegradation of Eumelanin and Pheomelanin and Its
Pathophysiological Implications. Photochem Photobiol 2018, 94, 409-420, doi:10.1111/php.12837.
46. Ozeki, H.; Ito, S.; Wakamatsu, K.; Thody, A.J. Spectrophotometric characterization of eumelanin and
pheomelanin in hair. Pigment Cell Res 1996, 9, 265-270, doi:10.1111/j.1600-0749.1996.tb00116.x.
47. Wakamatsu, K.; Nagao, A.; Watanabe, M.; Nakao, K.; Ito, S. Pheomelanogenesis is promoted at a weakly
acidic pH. Pigment Cell Melanoma Res 2017, 30, 372-377, doi:10.1111/pcmr.12587.
48. Smith, B.C. Group Wavenumbers and an Introduction to the Spectroscopy of Benzene Rings.
Spectroscopy 2016, 31, 34-37.
31
49. Smith, B.C. The C-O Bond, Part I: Introduction and the Infrared Spectroscopy of Alcohols. Spectroscopy
2016, 32, 14-21.
50. Smith, B.C. Organic Nitrogen Compounds, Part I: Introduction. Spectroscopy 2018, 34, 10-15.
51. Smith, B.C. The Carbonyl Group, Part V: Carboxylates - Coming Clean. Spectroscopy 2018, 33, 20-23.
52. Smith, B.C. The Carbonyl Group, Part I: Introduction. Spectroscopy 2017, 32, 31-36.
53. Smith, B.C. The C=O bond, Part III: Carboxylic Acids. Spectroscopy 2018, 33, 14-20.
54. Smith, B.C. Alcohols - The Rest of the Story. Spectroscopy 2017, 32, 19-23.
55. Smith, B.C. Organic Nitrogen Compounds V: Amine salts. Spectroscopy 2019, 34, 3037.
56. Ryu, J.H.; Messersmith, P.B.; Lee, H. Polydopamine Surface Chemistry: A Decade of Discovery. ACS
Applied Materials & Interfaces 2018, 10, 7523-7540, doi:10.1021/acsami.7b19865.
57. Lee, H.A.; Park, E.; Lee, H. Polydopamine and Its Derivative Surface Chemistry in Material Science: A
Focused Review for Studies at KAIST. Adv Mater 2020, 32, e1907505, doi:10.1002/adma.201907505.
58. Bongartz, V.; Brandt, L.; Gehrmann, M.L.; Zimmermann, B.F.; Schulze-Kaysers, N.; Schieber, A. Evidence
for the Formation of Benzacridine Derivatives in Alkaline-Treated Sunower Meal and Model Solutions.
Molecules 2016, 21, 91, doi:10.3390/molecules21010091.
59. Iacomino, M.; Weber, F.; Gleichenhagen, M.; Pistorio, V.; Panzella, L.; Pizzo, E.; Schieber, A.; d'Ischia, M.;
Napolitano, A. Stable Benzacridine Pigments by Oxidative Coupling of Chlorogenic Acid with Amino Acids
and Proteins: Toward Natural Product-Based Green Food Coloring. J Agric Food Chem 2017, 65, 6519-
6528, doi:10.1021/acs.jafc.7b00999.
60. Rzepecki, L.M.; Waite, J.H. A chromogenic assay for catecholoxidases based on the addition of l-proline
to quinones. Analytical Biochemistry 1989, 179, 375-381, doi:https://doi.org/10.1016/0003-
2697(89)90148-6.
61. Mavridi-Printezi, A.; Giordani, S.; Menichetti, A.; Mordini, D.; Zattoni, A.; Roda, B.; Ferrazzano, L.;
Reschiglian, P.; Marassi, V.; Montalti, M. The dual nature of biomimetic melanin. Nanoscale 2023, 16,
299-308, doi:10.1039/d3nr04696f.
62. Mavridi-Printezi, A.; Menichetti, A.; Ferrazzano, L.; Montalti, M. Reversible Supramolecular Noncovalent
Self-Assembly Determines the Optical Properties and the Formation of Melanin-like Nanoparticles. J
Phys Chem Lett 2022, 13, 9829-9833, doi:10.1021/acs.jpclett.2c02239.
63. Wang, X.; Kinziabulatova, L.; Bortoli, M.; Manickoth, A.; Barilla, M.A.; Huang, H.; Blancafort, L.; Kohler, B.;
Lumb, J.P. Indole-5,6-quinones display hallmark properties of eumelanin. Nat Chem 2023, 15, 787-793,
doi:10.1038/s41557-023-01175-4.
64. Battistella, C.; McCallum, N.C.; Gnanasekaran, K.; Zhou, X.; Caponetti, V.; Montalti, M.; Gianneschi, N.C.
Mimicking Natural Human Hair Pigmentation with Synthetic Melanin. ACS Central Science 2020, 6, 1179-
1188, doi:10.1021/acscentsci.0c00068.
65. Lampel, A.; McPhee, S.A.; Park, H.A.; Scott, G.G.; Humagain, S.; Hekstra, D.R.; Yoo, B.; Frederix, P.; Li,
T.D.; Abzalimov, R.R.; et al. Polymeric peptide pigments with sequence-encoded properties. Science
2017, 356, 1064-1068, doi:10.1126/science.aal5005.
66. Perna, G.; Lasalvia, M.; Capozzi, V. Vibrational spectroscopy of synthetic and natural eumelanin. Polym
Int 2016, 65, 1323-1330, doi:10.1002/pi.5182.
67. Bedran, Z.V.; Zhukov, S.S.; Abramov, P.A.; Tyurenkov, I.O.; Gorshunov, B.P.; Mostert, A.B.; Motovilov, K.A.
Water-Activated Semiquinone Formation and Carboxylic Acid Dissociation in Melanin Revealed by
Infrared Spectroscopy. Polymers (Basel) 2021, 13, doi:10.3390/polym13244403.
68. Micillo, R.; Panzella, L.; Iacomino, M.; Prampolini, G.; Cacelli, I.; Ferretti, A.; Crescenzi, O.; Koike, K.;
Napolitano, A.; d'Ischia, M. Eumelanin broadband absorption develops from aggregation-modulated
chromophore interactions under structural and redox control. Sci Rep 2017, 7, 41532,
doi:10.1038/srep41532.
32
