Ionization and tautomerism of methyl fluorescein and related dyes

Article abstract of DOI:10.1016/j.saa.2015.05.037

Abstract The protolytic equilibrium of methyl ether of fluorescein is studied in water, aqueous ethanol, and in other solvents. The constants of the two-step dissociation are determined by spectrophotometry. In water, the fractions of the zwitterionic, qu

Full text of DOI:10.1016/j.saa.2015.05.037

Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 150 (2015) 151–161
Contents lists available at ScienceDirect
Spectrochimica Acta Part A: Molecular and
Biomolecular Spectroscopy
journal homepage: www.elsevier.com/locate/saa
Ionization and tautomerism of methyl fluorescein and related dyes
a
a
b
Nikolay O. Mchedlov-Petrossyan a, , Tatyana A. Cheipesh , Sergey V. Shekhovtsov , Andrey N. Redko ,
⇑
Vladimir I. Rybachenko b, Irina V. Omelchenko c, Oleg V. Shishkin c,
a Department of Physical Chemistry, Kharkov V. Karazin National University, Kharkov 61022, Ukraine
b Institute of Physico-Organic Chemistry and Coal Chemistry, National Academy of Science of Ukraine, Donetsk 83114, Ukraine
c Institute for Single Crystals, National Academy of Science of Ukraine, Kharkov 61072, Ukraine
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
ꢀ
The molecular form of fluorescein
methyl ether in water is a mixture of
tautomers.
The fractions of the zwitter-ion,
quinonoid, and lactone are 11%, 6%,
and 83%.
In aqueous ethanol, the colorless
molecular lactone predominates.
The lactone structure of the
monoanion in solution is unlikely.
H3CO
O
O
OH
c(HCl),mol×L–1
0.63
0.40
0.16
:
paH* +
:
H2R+
R–
50
40
30
20
10
0
H2R+
8.46
7.91
7.55
7.36
7.11
6.90
6.80
6.27
5.82
COOH
ꢀ
pKa0
H3CO
paH* +
:
H3CO
O
OH
O
OH H3CO
O
1.31
1.51
1.69
O
HR
pKa1
ꢀ
ꢀ
COO
COOH
C
O
H3CO
O
O
H3CO
O
O
HR
O
C
R
COO
O
λ/nm
550
350
400
450
500
a r t i c l e i n f o
a b s t r a c t
Article history:
The protolytic equilibrium of methyl ether of fluorescein is studied in water, aqueous ethanol, and in
other solvents. The constants of the two-step dissociation are determined by spectrophotometry. In
water, the fractions of the zwitterionic, quinonoid, and lactonic tautomes are correspondingly 11%, 6%,
and 83%, as deduced from the UV–visible spectra. Corresponding study of the ionization of the methyl
ether ester of fluorescein, fluorescein ethyl ester, and sulfonefluorescein allows testing the correction
of the attribution of the microscopic dissociation constants of methoxy fluorescein. The results of nuclear
magnetic resonance and infrared spectroscopy, as well as the X-ray analysis confirm the predomination
of the lactonic structure of the molecular species in solid state and in DMSO. Contrary to it, the spectro-
scopic studies in both hydrogen-donor bond (HDB) and non-HBD solvents confirm that the presence of
lactonic monoanion is atypical for the dye under study and, with high probability, also for the mother
compound fluorescein.
Received 13 December 2014
Received in revised form 27 March 2015
Accepted 7 May 2015
Available online 20 May 2015
Keywords:
Fluorescein methyl ether
Molecular and nuclear magnetic resonance
spectroscopy
Dissociation constants
Tautomerism
X-ray structure analysis
Ó 2015 Elsevier B.V. All rights reserved.
Introduction
is the expanding applications of the abovementioned compounds
in versatile fields, including biochemistry [15–18].
Nowadays, much attention has been given to the structure,
absorption, and fluorescence of fluorescein dyes in solution [1–9].
The recent studies of the structure and spectra in the gas phase
are of special interest [10–14]. The evident reason of such activities
It should be mentioned, however, that the detailed ionization
scheme of the mother compound, fluorescein, in water [4,6,19–
23] and non-aqueous solvents was already studied in full and
reported in a set of publications [24–29].
In order to further this research we studied the protolytic equi-
librium of the monomethyl derivative of fluorescein, which may be
represented as a colorless lactone, 30-hydroxy-60-methoxyspiro[is
obenzofuran-1(3H),90-(9H)-xanthene]-3-one, or as a quinonoid,
2-(6-methoxy-3-oxo-3H-xanthene-9-yl) benzoic acid:
⇑
Corresponding author.
E-mail address: mchedlov@univer.kharkov.ua (N.O. Mchedlov-Petrossyan).
Deceased.
http://dx.doi.org/10.1016/j.saa.2015.05.037
1386-1425/Ó 2015 Elsevier B.V. All rights reserved.

152
N.O. Mchedlov-Petrossyan et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 150 (2015) 151–161
tautomeric equilibrium of methyl fluorescein. Two related dyes,
4
5
4
H3CO
O
O
5
sulfonefluorescein and ethyl ester of fluorescein, were also studied
in order to compare all the pk values with those of fluorescein in
the same mixed solvent.
H3CO
O
OH
7
2
7
2
6/
COOH
O
Also, the protolytic equilibrium of methyl fluorescein was stud-
ied in water, despite the limited solubility of the neutral molecular
6/
5/
5/
C
3/
species. This was necessary in order to compare the KT, K= , K== , and
O
3/
T
T
4/
4/
pk values with those for fluorescein, taking into account some col-
liding information concerning the tautomerism of fluorescein just
in water.
Indeed, Nagase et al. [38] proposed all the three tautomers,
whereas Scharf [39] and Zanker and Peter [40] considered the equi-
librium between H2L and H2Q. Hioka and colleagues trend to such
point of view in their recent paper [4].
At first glance, this compound is of less interest for the
physico-chemistry of xanthenes, because within the reasonable
pH range it cannot generate a double charged brightly emitting
anion. However, the examining of the ionic equilibrium of methyl
ether of fluorescein is necessary for better understanding the (partly
hidden) properties of the parent dye.
Using the visible spectra of neutral species, Lindqvist [19,41]
estimated the fractions of the tautomers: the ratio of H2Z , H2Q,
and H2L equals 2:1:5. We somewhat refined these approach; the
percentages of the zwitterionic, quinonoidal, and lactoid tautomers
was found to be 22%, 11%, and 67%, respectively [20].
Simultaneously a paper by Chen, Nakamura, and Tamura appeared
where only the lactoid tautomer, H2L, was presumed [32]. Later on,
Tamura and co-workers found the fractions of all the three tau-
tomers of fluorescein in water close to ours: 20%, 13%, and 67%,
respectively [21]. The corresponding values of 15, 15, and 70% were
then published by Klonis and Sawyer [22].
Alternatively, Fompeydie and Levillain [25,26] considered the
zwitterion rather as a kind of a transient form, whereas Diehl
et al. [42] assumed that the molecular species of fluorescein exist
in water just as this tautomer. However, Sjöback et al. [23] consid-
ered the quinonoid tautomer as ‘generally believed to be prevalent’
in aqueous solutions.
The detailed ionization equilibrium of fluorescein in solutions is
given in Scheme 1.
The structures possessing identical or similar chromophore sys-
tems are designed by the same letters, namely, H3Z+ and H2Z , H2Q
and HQꢁ, HXꢁ and X2ꢁ. Whereas the neutral form exists in solution
as a mixture of three tautomers, zwitterion H2Z , quinonoid H2Q,
and colorless lactone H2L, only the carboxylate tautomer, HQꢁ, is
typical for the monoanion. The phenolate tautomer HXꢁ appears
in small quantities only in pure non-hydrogen bond donor sol-
vents, such as DMSO, acetonitrile, and acetone [29], but it predom-
inates in the gas phase [10–14]. For the derivatives bearing halogen
atoms in the xanthene moiety, such as eosin, the zwitterionic tau-
tomer is less typical, whereas the monoanion exists predominantly
as HXꢁ [25,28].
The values of the tautomerization constants KT, K= , and K==, and
T
T
Hence, the verification of the problem using the methyl ether of
fluorescein as a model compound seems to be pertinent.
consequently the indices of the microscopic dissociation constants,
pk (Scheme 1), were estimated for fluorescein in water and in sev-
eral non-aqueous systems [6,7,20–22,28]. But for all that, some
structures were excluded from consideration. For example, the
anions–lactones HLꢁ and L2ꢁ (not shown in Scheme 1) were
regarded as less probable because in different solvents the varia-
tions of the maximal molar absorptivity of mono- and di-anions
were rather of solvatochromic nature and occurred simultaneously
with the shifts of the wavelength of the absorption band maxi-
mum. Such lactonic anions were registered only in the case of nitro
derivatives of fluorescein [30].
The above-mentioned regularities of tautomerism of fluorescein
dyes have been recently corroborated by quantum-chemical calcu-
lations [31].
The quantitative study of ionization and tautomerism of the
methyl ether of fluorescein and several related dyes allows one
to verify the assumptions used earlier and compare the pk values
with those of fluorescein. The probable detailed protolytic equilib-
rium of methyl fluorescein is given in Scheme 2.
Experimental
Materials
Solvents for synthesis were purified according to standard
methods. Solvents for visible spectroscopic measurements and
pKa determination were of analytical and spectroscopic grade.
Buffer solutions components, i.e., phosphoric, acetic, hydrochloric
acids, as well as sodium chloride were of analytical grade.
Sodium hydroxide solution was prepared using CO2-free water
and kept protected from the CO2-containing air. 1,8-Diazabicycl
o[5.4.0]undec-7-ene, or DBU (Merck), was used as commercially
obtained.
Synthesis of the dyes
The probability of existence of the lactonic monoanion should
be verified more directly than in the case of fluorescein, because
here the Rꢁ species predominates within a broad pH range. On
the other hand, such a study allows elucidating how replacing of
OH by OCH3 influences the acidity of the COOH and the remaining
OH group.
Some data on ethers and esters of fluorescein dyes are available
in the literature [25–37]; Amat-Guerri et al. studied the ethers of
eosin and rose bengal B and thus estimated the fractions of the lac-
tonic monoanions of these dyes [37].
In this paper, we report the characterization of methyl ether of
fluorescein and its methyl ester in solid (X-ray, IR spectra) and liq-
uid (IR, 1H NMR, and 13C NMR spectroscopy) states, the dissocia-
tion constants of the dyes in 50 mass% aqueous ethanol by
means of vis-spectroscopy, and the results of examination of the
Methyl ether ester of fluorescein
This compound was synthesized following Fischer and Hepp
[43]. The product was recrystallized from the CCl4/CHCl3 (3:1 by
volume) mixture: 2.75 g was dissolved in 40 mL of the mixed sol-
vent under heating. Then the solution was filtered and cooled to
give the orange precipitate (1.9 g). 1H NMR ((CD3)2S@O) d/ppm:
8.19 (1H, dd, J = 7.1, J = 2.1, 30-H), 7.94–7.68 (2H, m, J = 7.1, J = 2.1,
40,50-H), 7.47 (1H, dd, J = 7.1, J = 2.1, 60-H), 7.21 (1H, d, J = 2.1,
4-H), 6.95–6.68 (3H, m, 1,2,8-H), 6.36 (1H, dd, J = 9.5, J = 2.1,
7-H), 6.21 (1H, d, J = 2.1, 5-H), 3.88 (3H, s, Ph-O-CH3), 3.55 (3H, s,
Ph-CO-O-CH3). 13C NMR ((CD3)2S@O) d/ppm: = 183.89, 165.21,
163.92, 158.39, 153.60, 150.14, 133.92, 133.24, 130.73, 130.40,
130.08, 129.39, 129.51, 128.88, 116.65, 113.60, 114.31, 104.60,
100.60. IR/cmꢁ1, selected bands: 1726, 1642, 1587, 1509, 1453,
1256, 1211, 1105.

N.O. Mchedlov-Petrossyan et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 150 (2015) 151–161
153
HO
O
OH
H3R+
COOH
H3Z+
Ka0
k ,COOH
k0,OH
HO
O
OH
HO
O
O
HO
O
OH
KT
H2R
O
C
O
COO
COOH
K'T
H2L
H2Z
H2Q
Ka1
k1,Z
k1,OH
k1,COOH
O
O
O
HO
O
O
KT
x
HR
COOH
COO
HQ
HX
Ka2
k2,OH
k2,COOH
O
O
O
R2
COO
X2
Scheme 1. Protolytic equilibrium of fluorescein: KT ¼ ½H2Lꢂ=½H Qꢂ; K=T ¼ ½H2Zꢃꢂ=½H2Qꢂ; K=T= ¼ KT=K= ¼ ½H2Lꢂ=½H Zꢃꢂ; KT ¼ ½HXꢁꢂ=½HQꢁꢂ;
k
=
aH aH =aH
þ
3Zþ
;
2
T
2
2Zꢃ
ꢃ;COOH
x
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
þ
þ
þ
þ
þ
þ
3Zþ
2 Zꢃ
k0;OH ¼ aH aH Q =aH ; k1;Z ¼ aH aHQ =aH ; k1;COOH ¼ aH aHQ =aH Q ; k1;OH ¼ aH aHX =aH Q ; k2;OHaH aX2ꢁ =aHQ ; k2;COOH ¼ aH aX2ꢁ =aHX
.
2
2
2
Methyl ether of fluorescein
8.08 (1H, d, J = 4, 30-H), 7.58 (2H, m, J = 4, 40,50-H), 7.17 (1H, d,
J = 4, 60-H), 7.08 (1H, s, 5-H), 6.84 (1H, d, J = 8, 8-H), 6.80 (1H, d,
J = 8, 7-H), 6.74 (1H, d, J = 8, 1-H), 6.32 (1H, d, J = 8, 2-H), 6.27
(1H, s, 4-H), 3.86 (3H, s, Ph-O-CH3). 13C NMR ((CD3)2S@O)
d/ppm = 178.55, 168.79, 162.82, 157.15, 153.15, 138.02, 137.15,
130.74, 130.48, 129.66, 129.07, 128.95, 127.66, 124.56, 114.51,
113.27, 112.70, 103.52, 100.31, 56.08.
According to the procedure described by Fischer and Hepp [43],
1.54 g of the above methyl ether ester in 15 mL of methanol and
0.8 g potassium hydroxide (1 mol Lꢁ1) were heated under reflux
for 2 h. After cooling of reaction mixture, 100 mL of distilled water
was added. The transparent solution thus obtained was acidified
by adding HCl. The resulting yellow flocky precipitate was filtered
off, washed with distilled water and dried. This unpurified product
(1.2 g) was recrystallized from ethanol–water mixture (50 vol%).
The crystals thus obtained were used for X-ray analysis. 1H NMR
((CD3)2S@O) d/ppm: 10.14 (1H, s, Ph-OH), 7.98 (1H, d, J = 7.2,
30-H), 7.83–7.64 (2H, m, J = 7.2, 40,50-H), 7.25 (1H, d, J = 7.2, 60-H),
6.91 (1H, d, J = 2.1, 4-H), 6.73–6.51 (5H, m, 1,2,5,7,8-H), 3.79 (3H, s,
Ph-O-CH3). 13C NMR ((CD3)2S@O) d/ppm = 168.68, 161.03, 159.55,
152.50, 151.88, 151.77, 135.67, 130.16, 129.11, 128.96, 126.03,
124.67, 124.01, 112.79, 111.91, 110.99, 109.44, 102.2, 100.78,
82.71, 55.66. IR/cmꢁ1, selected bands: 1718, 1609, 1508, 1430,
1286, 1257, 1171, 1120.
Sulfonefluorescein
14 g of P4O10 was carefully added to 10 mL of 85 mass% aqueous
phosphoric acid. The stirred mixture was heated to 100 °C to give
transparent solution. 5.0 g of ammonium salt of the
2-sulfobenzoic acid were slowly dissolved in the reaction mixture
at 150 °C. Then 5.0 g resorcinol was gradually added, and the
formed dark red product was heated to 170–190 °C for 3 h. Along
with the progress of reaction, the target product precipitates in
form of small glossy purple crystals, and the mass becomes more
viscous. The reaction mixture was cooled to 80–90 °C and diluted
with 150 mL ethanol–water mixture (50 vol%). After filtering off,
washing with ethanol (95.6 mass%) and drying, 6.29 g of unpurified
precipitate was obtained. As sulfonefluorescein is relatively poor
soluble in the most of readily accessible solvents, the
The sodium salt was prepared by addition of NaHCO3 to the
aqueous suspension of methyl ether of fluorescein. After evapora-
tion of water from the solution obtained, the target product was
extracted by 2-propanol and dried. 1H NMR ((CD3)2S@O) d/ppm:

154
N.O. Mchedlov-Petrossyan et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 150 (2015) 151–161
H3CO
O
OH
H2R+
COOH
H2Z+
Ka0
k ,COOH
H3CO
k0,OH
H3CO
O
OH
O
O
H3CO
O
OH
KT
HR
O
C
O
COO
COOH
K'T
HL
HZ
HQ
Ka1
k1,Z
k1,COOH
H3CO
O
O
H3CO
O
O
R
O
C
O
COO
Q
L
Scheme 2. Protolytic equilibrium of methyl fluorescein: KT ¼ ½HLꢂ=½HQꢂ; K=T ¼ ½H2Zꢃꢂ=½H2Qꢂ; K=T= ¼ KT=K= ¼ ½HLꢂ=½HZꢃꢂ; KL ¼ ½Lꢁꢂ=½Qꢁꢂ;
k
¼ aH aHZꢃ =aH
þ
2 Zþ
;
ꢃ;COOH
T
ꢁ
ꢁ
þ
þ
þ
2 Zþ
2Zꢃ
k0;OH ¼ aH
a
HQ =aH ; k1;Z ¼ aH aHQ =aH ; k1;COOH ¼ aH aQ =aHQ
.
recrystallization was carried thought conversion of the compound
into its soluble disodium salt and subsequent acidification of the
solution by HCl [44]. The solution of the disodium salt (1 g) in
1000 mL of water was heated to boiling, then acidified by appropri-
ate amount of HCl, and slowly cooled. This results in the precipitat-
ing of sulfonefluorescein in form of large purple crystals. The
ethanol–water mixture (50 vol%) could also be used as a solvent
instead of water (1 g salt per 50–100 mL). 1H NMR ((CD3)2S@O)
d/ppm: 7.99 (1H, d, J = 7.4, 31-H), 7.68 (1H, t, J = 7.4, 51-H), 7.58
(1H, t, J = 7.4, 41-H), 7.37 (2H, d, J = 8.8, 1,8-H), 7.31–7.21 (3H, m,
61,4,5-H), 7.13 (2H, d, J = 8.8, 2,7-H). IR/cmꢁ1, selected bands:
1573, 1455, 1302, 1213, 1136, 1118, 1041.
Spectrophotometer (Bruker) with Alpha-Platinum ATR module
and either in KBr pellets or in DMSO solutions with application
of FTIR spectrometer SPECTRUM ONE (PerkinElmer) in the range
of 400–4000 cmꢁ1 by Dr. D. S. Sofronov, Division of Functional
Materials Chemistry, Institute for Single Crystals, National
Academy of Science of Ukraine.
X-ray diffraction study: crystals of monomethyl fluorescein
ꢀ
(C21H14O5,
Mr = 346.32)
are
triclinic,
P1,
a = 7.956(2),
b = 10.813(3), c = 10.831(2) Å,
a
= 94.84(2)°, b = 108.52(2)°,
Z = 2,
dcalc = 1.402 g smꢁ1
= 0.101 mmꢁ1, F(000) = 360. 7051 reflections (3750 indepen-
c
= 108.44(2)°,
V = 820.3(3) Å3,
,
l
dent, Rint = 0.106) were collected on an ‘‘Xcalibur-3’’ diffractometer
(MoK radiation, CCD-detector, graphite monochromator,
-scanning, 2hmax = 50°). Structure was solved by direct methods
The collection of the NMR and IR spectra is available in the
Supplementary data.
a
x
and refined against F2 within anisotropic approximation for all
non-hydrogen atoms by full-matrix least squares procedure using
OLEX2 program package [47] with SHELXS and SHELXL modules
[48]. All H atoms were placed in idealized positions (C–H = 0.93–
0.96 Å, O–H = 0.82 Å) and constrained to ride on their parent
atoms, with Uiso = 1.2Ueq (except Uiso = 1.5Ueq for methyl and
hydroxyl groups). Final refinement was converged at wR2 = 0.092
for all 2861 reflections (R1 = 0.060 for 899 reflections with
Apparatus
The visible absorption spectra were run on a Hitachi U-2000
spectrophotometer. Fluorescence spectra were measured using
Hitachi 850 apparatus. The pH determinations were performed
using a potentiometer R 37–1 and a pH-meter pH-121, equipped
with an ESL-63-07 glass electrode and an Ag|AgCl reference elec-
trode, in a cell with liquid junction (1.00 mol Lꢁ1 KCl). The glass
electrode calibration was performed with standard aqueous buffer
solutions (pH 1.68, 4.01, 6.86, and 9.18 at 25.0 °C). In aqueous etha-
nol, the proton activity aꢄHþ was standardized to infinite dilution in
the given solvent, the experimental (instrumental) pH values were
corrected for the diffusion potential and proton solvation by rela-
tion: paꢄHþ ¼ pHinstr ꢁ 0:20 [45,46]. The 1H NMR spectra were
recorded on Mercury Varian VX-200 spectrometer at 200 MHz
and Bruker Avance II 400 at 400 MHz, the 13C NMR spectra were
recorded on Bruker Avance II 400 spectrometer at 100 MHz. The
IR spectra were obtained by Dr. D. Yu. Filatov on Alpha FT-IR
F > 4r(F), S = 0.78).
Atom coordinates and crystallographic parameters have been
deposited to the Cambridge Crystallographic Data Centre (CCDC
1026934). These data can be obtained free of charge from the
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.
uk/data_request/cif.
Procedure
For pKa determination, the stock solutions were prepared from
weighted amount of dyes using 95.6 mass% ethanol as a solvent.

N.O. Mchedlov-Petrossyan et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 150 (2015) 151–161
155
When preparing the initial solution of fluorescein methyl ether in
aqueous solutions, small amount of NaOH solution was added for
enhancing the dye solubility. In solutions of methyl ether ester of
fluorescein and ethyl ester of fluorescein in 50 mass% C2H5OH,
the paꢄHþ values did not exceed 11 in order to avoid the hydrolysis
of the ester groups. The working solutions were prepared by mix-
ing required volumes of the stock dye solution, aqueous buffer
components and NaCl solutions with 95.6 mass % C2H5OH.
Working dyes concentrations were ranged from 6 ꢅ 10ꢁ6 mol Lꢁ1
to 2.6 ꢅ 10ꢁ4 mol Lꢁ1. Complete conversion of the dyes into the
corresponding ultimate basic and acidic form was attained in the
presence of 1 ꢅ 10ꢁ4 mol Lꢁ1 NaOH and 2 mol Lꢁ1 HCl respectively.
The sets of 15–20 (for dyes dissociating in two steps) and 6 (for
methyl ether ester of fluorescein) working solutions at different
paꢄHþ (or pH) were utilized. The value of ionic strength was main-
tained constant (0.05 mol Lꢁ1) by adding a proper amount of
NaCl to the buffer components or to dilute NaOH and HCl solutions,
except the experiments with high HCl concentrations.
The pKa values were calculated jointly with the spectra of the
neutral (for methyl ether and ethyl ester of fluorescein) or anionic
(for sulfonefluorescein) species by the CLINP program [49]. The
iterative procedure used was based on the minimizing the
Huber’s criterion function by the Gauss–Newton method. As a rule,
the wavelengths around the absorbance maxima were considered
as the most informative analytical positions.
Fig. 1. Molecular structure of the monomethyl fluorescein according to the X-ray
diffraction data. All non-hydrogen atoms are shown as 40% thermal ellipsoids.
fluorescein structural analogues (mean deviation of the corre-
sponding atom is 0.17 Å [53]). Xanthene and isobenzofuran moi-
eties are almost orthogonal (angle between planes is 88.6(8)°
that is equal to mean value for its structural analogues [53]).
That causes presence of shortened intramolecular contacts
C(15)ꢆ ꢆ ꢆH(14) 2.68 Å and C(15)ꢆ ꢆ ꢆH(7) 2.69 Å (sum of VdW radii
is 2.87 Å [56]). In the crystal, molecules of monomethyl fluorescein
form dimers due to intermolecular hydrogen bonds O(3)–H(3)ꢆ ꢆ ꢆ
O(5)0 (1 ꢁ x, ꢁy, 2 ꢁ z; Hꢆ ꢆ ꢆO0 2.18 Å, O–Hꢆ ꢆ ꢆO 146°).
The relative permittivity of the 50 mass% aqueous ethanol is
equal to 49.0. The ionic activity coefficients of the ionic dye species
were calculated using the Debye–Hückel equation (second
approach, ionic parameter = 0.5 nm) to estimate the thermody-
namic pKa values. The corresponding corrections are +0.16, +0.47,
and ꢁ0.16 for pKa1, pKa2, and pKa0 respectively.
The solutions for measuring the absorption spectra of the
methyl ether anion and methyl ether ester neutral species in
non-aqueous solvents were prepared by dilution of the stock solu-
tions of the dyes either in acetone or in 95.6 mass% ethanol.
The intensive band 1718 cmꢁ1 in the IR spectrum of the solid
methyl ether lactone of fluorescein should unambiguously be
ascribed to the stretching vibrations of the C@O group of c-lactone
cycle. The X-ray data confirm that the sample under consideration
exists without any solvent molecule. This is a remarkable observa-
tion, because the above cited authors state that the lactone of the
parent compound can be prepared in solid state only in form of a
Results and discussion
Characterization of the samples
solvate [51]. They report the value
m
ðC@OÞ ¼ 1750 cmꢁ1 for the
In the crystal, monomethyl fluorescein exists in the neutral lac-
tone form (Fig. 1). The C(1)–O(4) bond length of 1.535(6) Å is
rather long as compared to fluorescein and its structural analogues
(mean value is 1.508 Å [50–53]). Also it is much longer than typical
C–O bond length in common
with shortening of C(1)–C(2) and C(5)–O(2) bonds (1.497(6) Å
and 1.344(6) Å, mean values are 1.527 Å and 1.370 Å, respectively
[54]) that indicates presence of intramolecular
the methylresorcin moiety. This conjugation interaction does not
affect the C(1) atom geometry, it retains tetrahedral configuration
(valence angles are 101.0(3)–115.2(4)°).
Particularly interesting is that hydroxyl group, unlike methoxy
group, does not take part in such conjugation: the C(1)–C(9) bond
lengths is close to the mean value, and C(12)–O(3) bond of
1.388(6) Å is even longer than Car–OH bonds in fluorescein
(1.352–1.364 Å according to [50–52]). Values of C(3)–O(1) and
C(10)–O(1) bond lengths (1.393(6) Å and 1.378(7) Å) that are close
to mean value for diaryl ethers (1.384 Å [54]) indicate that O(1)
atom also does not take part in conjugation interaction in the
molecule, as it was supposed previously for fluorescein analogues
in solid state [55].
complexes of fluorescein lactone with acetone or methanol.
On the other hand, Markuszewsky and Diehl determined the
value 1730 cmꢁ1 for the C@O group of the fluorescein lactone, with
a shoulder at 1760–1770 cmꢁ1 which they attributed to the lactone
complex with 1,4-dioxane [48]. In earlier publications, the band at
1729–1730 cmꢁ1 was reported [29,57].
c-lactones 1.464 Å [54]. Together
The value 1750 cmꢁ1 was registered by us for the molecular
species of fluorescein in DMSO and CHCl3 solutions [58,59]. It
should be concluded that solvation of the lactone cycle results in
p-conjugation in
20–40 cmꢁ1 increase in the
m
ðC@OÞ value.
In the IR spectrum of methyl ether of fluorescein in DMSO, the
band
m
ðC@OÞ ¼ 1756 cmꢁ1 was registered, in line with the lactone
structure, HL. The signal d = 82.71 ppm registered by us in the 13C
NMR spectrum in DMSO-d6 may be taken as a very strong support
for the prevalence of the lactonic structure. It should be ascribed
unequivocally to the central carbon atom of the last-named tau-
tomer that is in the state of sp3-hybridization; this value coincide
with the published d = 83.21 value for the 30,60-dimethoxyfluorane
in CDCl3 [60]. Meanwhile, the signal for the neutral species of the
methyl ether ester
Q is shifted towards the downfield,
d = 150.14 ppm, in accordance with the sp2-hybridization.
On the other hand, very narrow shape of the 10.1 ppm signal in
the 1H NMR spectrum (see Supplementary data) gives evidence of
the lack of any expressed specific interactions between DMSO and
Xanthene moiety is almost planar: two benzene rings are
slightly rotated around each other, angle between planes is 6.2°.
The C(1) atom deviates from the mean plane of the rest xanthene
atoms on 0.22 Å (plane accuracy is 0.06 Å). That is typical for

156
N.O. Mchedlov-Petrossyan et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 150 (2015) 151–161
the OH group of the HL species. This is in line with the weak acidity
of the hydroxyl group of fluorescein dyes in DMSO [27,28,58].
The intensive bands 1726 cmꢁ1 and 1642 cmꢁ1 in the IR spec-
trum of the solid molecular form of methyl ether ester of fluores-
H2R+
50
40
30
20
10
0
paH* +
:
c(HCl),mol×L–1
0.63
0.40
0.16
:
8.46
7.91
7.55
7.36
7.11
6.90
6.80
6.27
5.82
cein should be attributed to
m
ðC@OÞ of the COOCH3 group and of
the quinone in the xanthene moiety respectively. Fompeydie and
Levillain reported the value of 1645 cmꢁ1 for their ‘non-lactone’
sample, obtained from aqueous solution [26] (probably, quinonoid
H2Q). We registered the bands 1637, 1612, and 1642 cmꢁ1 for the
molecular forms of 6-hydroxy-9-phenyl fluorone, n-decyl ester of
fluorescein, and ethyl ether ester of fluorescein, and 1720–
1722 cmꢁ1 for the last two dyes possessing the COOAlk groups.
R–
paH* +
:
1.31
1.51
1.69
1.87
Dissociation constants in 50 mass% aqueous ethanol
HR
The numbering of the stepwise dissociation constants is as fol-
lows: the constant Kað1ꢁzÞ corresponds to the step
zꢁ1
HjRz ¢ H
R
þ Hþ. Therefore, the constants determined for the
jꢁ1
fluorescein methyl ether describe the equilibria (1) and (2):
350
400
450
500
550
λ/ nm
H2Rþ ¢ HR þ Hþ; Ka0
ð1Þ
ð2Þ
Fig. 2. Absorption spectra of fluorescein methyl ether in 50 mass% aqueous ethanol
at different paꢄHþ and limiting dye forms.
HR ¢ Rꢁ þ Hþ; Ka1
The absorption spectra at different pH values and the depen-
dences of molar absorptivities vs. pH are exemplified in Figs. 2–6.
For fluorescein the cation H3R+ is a triprotic acid, and the third
40
dissociation step is HRꢁ ¢ R2ꢁ þ Hþ, with the equilibrium constant
Ka2 (Scheme 1). Sulfonefluorescein is also a triprotic acid, but the
cation exists at rather low pH values and the Ka0 was not deter-
mined here:
35
30
1
480 nm
480 nm
1'
25
20
15
10
5
440 nm
2
435 nm
2'
The ethyl ester of fluorescein and the 6-hydroxy-9-phenylfluorone
are dyes with ‘‘cut off’’ carboxyl group, and they are to be described
via equilibria (1) and (2), but one should keep in mind that in this
case it deals about the successively dissociation of OH groups: that
of the cation (Ka0) and of the quinonoid molecule (Ka1):
0
0
2
4
6
8
10
paH* + (pH)
Fig. 3. Molar absorptivities of fluorescein methyl ether versus pH or paꢄHþ in
aqueous (1, 10) and 50 mass% aqueous ethanol (2, 20) solutions, respectively.
_
HO
O
+
OH
HO
O
O
O
O
O
H3CO
O
+
OH
H3CO
O
O
COOC2H5
COOC2H5
COOC2H5
+
+
H
COOCH3
COOCH3
_
HO
O
+
OH
HO
O
O
O
O
O
HZ+
Q
The pKa data obtained are presented in Table 1; the main spectral
data are compiled in Table 2. It should be emphasized that the spec-
tra of the individual ionic and molecular forms (bold lines) have
been singled out from the initial data during the calculation process.
For fluorescein methyl ether ester in 50 vol% CH3OH, S. Niizuma
et al. [62] reported pKa0 ¼ 2:65. As the pKa0 values for similar
H2Z+
HQ
X
Finally, for methyl ether ester, only the first step (HRþ ¢ R þ Hþ) is
possible:

N.O. Mchedlov-Petrossyan et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 150 (2015) 151–161
157
100
90
80
70
60
50
40
30
20
10
0
Methyl fluorescein in water
During the study of fluorescein ether in water, we met difficul-
ties caused by the rather limited solubility of the neutral (molecu-
lar) species. The initial solution was prepared in ethanol-free
water, with addition of small amount of NaOH solution. In the
working aqueous solutions, a slow decrease in absorbance, up to
complete decoloration after 3 – 4 days, was observed within the
pH range from 4.90 to 2.87. However, over 30 min no appreciable
changes in absorbance of freshly prepared solutions occur. Thus,
some of the results were obtained with oversaturated solutions.
The pKa values are presented in Table 3; the data for the related
dyes obtained earlier are compiled ibidem (Fig. 7).
R2–
paH* +
:
H2R
paH* +
:
7.64
7.43
7.28
7.04
6.93
6.61
6.35
5.73
2.08
2.67
2.87
3.12
3.27
3.60
HR–
The kmax=nm (emax ꢅ 10ꢁ3=L molꢁ1 cmꢁ1) for methyl fluorescein
in water are as follows: Rꢁ: 454 (31.9) and 474 (30.7); HR: 433
(7.6); H2R+: 437 (54.0). Naturally, the analogue of the X2ꢁ species
of fluorescein (Scheme 1, R2ꢁ = X2ꢁ), possessing the band with
kmax ¼ 490:5 nm (emax = 88.0 ꢅ103) is absent among the species
of its methyl ether (Scheme 2). But the band of the cationic form
H2R+ (i.e., H2Z+) coincides with that of the parent dye, H3Z+:
λ/ nm
400
450
500
550
437 nm (emax = 54.3 ꢅ103). The similarity of the above Rꢁ spectra
Fig. 4. Absorption spectra of sulfonefluorescein in 50 mass% aqueous ethanol at
different paꢄHþ and limiting dye forms (heavy lines).
of methyl fluorescein and that of HRꢁ of fluorescein [the HQꢁ tau-
tomer; kmax ¼ 454—474 nm; emax ¼ ð32:7—33:8Þ ꢅ 103] give evi-
dence for the structure of Qꢁ type. The fluorescence spectrum of
this species in water was also measured; the quantum yield was
equilibria (dissociation of the cations of 6-hydroxy-9-phenyl fluo-
rone and fluorescein ethyl ester) decrease on going from water
(pKa0 around 3.0, see below) to water-organic mixtures, this result
agrees with our one, because 50 vol% of methanol corresponds to ꢇ
43 mass%.
estimated as
u
¼ 0:40, referring to the
u
¼ 0:93 value of the fluo-
rescein dianion R2ꢁ as standard. Note, that thus obtained value is in
line with those reported for the HRꢁ monoanion of fluorescein in
Several data in aqueous alcohols are available in the literature
for fluorescein. So, Frolov et al. [63] potentiometrically determined
in aqueous 52.7 mass% 2-propanol the thermodynamic values
pKa1 ¼ 6:90 and pKa2 ¼ 7:78 (at ionic strength 0.1 mol Lꢁ1, the val-
ues are 6.62 and 7.44). Diehl et al. [42] obtained via the same
method the values 6.38 and 7.16 in 50% aqueous ethanol at
0.1 mol Lꢁ1 (NaCl). Fompeydie and Levillain determined the values
pKa1 ¼ 5:60 and pKa2 ¼ 6:86 spectrophotometrically in 50 vol%
aqueous methanol (ionic strength not indicated); under the same
conditions, the value for ethyl fluorescein: pKa1 ¼ 6:36 [25].
For fluorescein methyl ether ester in 50 vol% CH3OH, Niizuma
et al. [62] reported kmax = 444 nm and 458 nm for the cation and
neutral form respectively.
water:
u
¼ 0:36 [22] and 0.37 [23]. The single-charged anion of
fluorescein contains the OH group in the xanthene moiety (the
HQꢁ tautomer in Scheme 1), instead of the CH3O group in the case
of the Qꢁ anion of methyl ether. Finally, the relatively low emax
values of the HR species of methyl ether permits to assume the
substantial fraction of the colorless lactone HL; for the H2R of the
parent dye, the emax values 13.9 ꢅ103 at 437 nm and (3–4) ꢅ103
at k within the range of 470–485 nm allowed to deduce the
co-existence of the colored tautomers H2Z and H2Q with the
colorless H2L [20]. (The data for fluorescein in water are from
Ref. [20].)
0.7
80
HR+
R–
70
0.6
0.5
0.4
paH* +
:
paH* +
60
50
40
30
20
10
0
:
1.95
2.16
2.37
2.64
2.82
3.02
7.50
7.17
6.91
6.71
6.35
6.21
5.76
H2R+
HR
paH* +
:
R
3.12
2.69
2.40
2.15
1.83
1.54
0.3
0.2
0.1
0
λ/ nm
400
450
500
550
λ/ nm
400
450
500
550
Fig. 5. Absorption spectra of fluorescein ethyl ester in 50 mass% aqueous ethanol at
different paꢄHþ and limiting dye forms (heavy lines).
Fig. 6. Absorption spectra of fluorescein methyl ether ester in 50 mass% aqueous
ethanol at different paꢄHþ and limiting dye forms (heavy lines).

158
N.O. Mchedlov-Petrossyan et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 150 (2015) 151–161
Table 1
Table 3
The thermodynamic values of the indices of dissociation constants in 50 mass%
aqueous ethanol.
The thermodynamic values of the indices of dissociation constants in water.
Compound
pKa0
pKa1
pKa2
Compound
pKa0
pKa1
pKa2
Methyl ether of fluorescein
Ethyl ester of fluoresceina
6-Hydroxy-9-phenyl fluoroneb
Sulfonefluoresceinc
1.94 0.01
2.94 0.07
3.10 0.02
–
4.73 0.01
6.31 0.03
6.28 0.06
3.22 0.11
4.45 0.02
–
–
–
Methyl ether ester of fluorescein
6-Hydroxy-9-phenyl fluoronea
Ethyl ester of fluorescein
Sulfonefluorescein
2.28 0.01
2.34 0.03
2.05 0.01
–
0.39 0.01
0.94 0.05
–
–
–
–
6.80 0.08
6.71 0.01
3.22 0.01
7.32 0.01
6.82 0.05
6.76 0.03
6.80 0.01
7.63 0.01
–
7.66 0.05
Fluoresceind
2.14 0.01
Methyl ether of fluorescein
a
b
c
Fluoresceinb
Ref. [64].
Refs. [20,61,64].
Refs. [27,58].
Refs. [6,20,27,28].
a
Ref. [61].
Ref. [64].
b
d
Tautomerism of the neutral form of the fluorescein methyl ether
In water solution, the molecular spectrum (Fig. 8) gives evi-
dence for the presence both of the zwitterion HZ and quinonoid
HQ. Again, the band 433 nm is hypsochromically shifted as com-
pared with the cationic band of this dye in water (437 nm), thus
confirming the influence of the COOꢁ group on the xanthene
portion.
The shape of the spectral curves of the methyl ether and methyl
ether–ester obtained in the present paper agrees with those pub-
lished by others [32–36]. It should be noted, that the presence of
the charged COOꢁ group in the phthalic residue shifts the band
of the xanthene moiety hypsochromically [7,22,25,27,28,37,
58,64,65]. This finding has been supported via quantum-chemical
calculations [66]. As it is seen from Table 2, the band of the dianion
of fluorescein, structure X2ꢁ, is shifted by 8–10 nm towards the lower
wavelengths against the bands of 6-hydroxy-9-phenylfluorone and
ethyl ester of fluorescein (structures Xꢁ).
The comparison of the kmax values of the methyl ether anion,
structure Qꢁ (Scheme 2), and fluorescein monoanion (HQꢁ), with
those of the molecular species of ether–ester (structure Q) and
6-hydroxy-9-phenylfluorone and ethyl ester of fluorescein (HQ)
also confirm this regularity (Table 1). The emax values are very close
for Qꢁ and Q species of methyl ether and the ether-ester, respec-
tively. However, on going from 50% to 96% ethanol or entire
organic solvents, the molar absorptivities exhibit some differences.
The absorption spectrum of the methyl ether molecular form
HR in 50% ethanol resembles by shape that of the anion of the same
dye (Figs. 2 and 8). Here, the band of the HQ tautomer is also
batochromically shifted against the band of the anion Qꢁ, whereas
the tremendous difference in intensity is certainly caused by the
high fraction of the colorless tautomer HL.
The fractions of the tautomers HZ and HQ in water are 0.11 and
0.06 respectively, and the rest refer to the lactone HL (0.83).
Consequently, KT ¼ 13:8; K=T ¼ 1:83. In 50% ethanol, the zwitteri-
onic tautomer does not manifest itself in the spectrum of the
molecular species, whereas the fractions of the quinonoid and lac-
tone are 0.01 and 0.99; KT ¼ 99. The corresponding fractions for
the parent compound fluorescein are in water [6,20,27,28]: 0.22
(H2Z ), 0.11 (H2Q), and 0.67 (H2L), while in 50 mass% ethanol
[64]: 0.0325 and 0.967 (KT ¼ 29:7). One may note that the ten-
dency to lactone formation is somewhat more expressed in the
case of the fluorescein methyl ether.
At this place it should be noted that all these calculations, as
well as the above cited for the parent compound fluorescein, are
made with the connivance about the absence of the colorless tau-
tomer of the monoanion (Scheme 2). Therefore, it was necessary to
examine this statement.
Examining the possibility of lactone–monoanion formation
The fractions of the tautomers of the neutral form HR may be
One of the aims of the present study was to verify the absence
of the monoanion-lactone of the methyl ether and thus to conclude
about the corresponding tautomerism of the fluorescein
(Scheme 2, Qꢁ ¢ Lꢁ). The intensity of the methyl ether anionic
band in water and in 50% aqueous ethanol stays practically con-
stant. For estimating the molar absorptivities of fluorescein methyl
ether monoanion in methanol, 95.6% aqueous ethanol, 1-butanol,
dichloromethane, and acetone (Fig. 9), the working solutions were
prepared by diluting of initial dye solution in acetone with appro-
priate solvent. The volume fraction of acetone in target mixture
equals 1%. In water and aqueous ethanol, the complete
estimated using the relation
e
¼
eHZꢃ
ꢅ
aHZꢃ
þ
eHQ
ꢅ
aHQ at a fixed
HQ . As it
wavelength, with understanding that aHL ¼ 1 ꢁ aHZꢃ
ꢁ
a
was earlier assumed for fluorescein [6,7,19,20,58], the absorptivi-
ties of the zwitterionic and quinonoidal tautomers, especially
around the band maxima (Fig. 8), may be equated to those of the
cation and monoanion, respectively. Therefore, before using the
ꢁ
2Zþ
molar absorptivities eH and eQ instead of eHZꢃ and
eHQ , respec-
tively, the spectra of the ionic forms were shifted to several
nanometers in order to reach the coincidence of their maxima with
those of the corresponding tautomers.
Table 2
Spectral characteristics of dyes in 50 mass % aqueous ethanol.
Dye
kmax/nm (
Dianion
–
e
ꢆ 10ꢁ3/L molꢁ1 cmꢁ1
)
Monoanion
Neutral form
Cation
Fluorescein methyl ether (in water)
Fluorescein methyl ether
452–456 (30.9)
472–476 (29.8)
452–455 (30.9)
478–481 (27.4)
–
432–434 (7.61)
436–438 (54.0)
–
454–457 (0.33)
481–488 (0.26)
457–459 (29.1), 486 (24.5)
457, 485
456–459, 483–487
445–447 (54.6)
441 (54.5)
Fluorescein methyl ether ester
6-Hydroxy-9-phenylfluorone
Fluorescein ethyl ester
–
–
–
442–443 (56.9)
503
504–505
445
445
–
Sulfonefluorescein
503 (91.6)
455–458 (30.7)
483–487 (28.3)
455 (34.3)
Fluoresceina
Ref. [64].
495 (88.5)
455 (0.978)
445 (62.4)
a

N.O. Mchedlov-Petrossyan et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 150 (2015) 151–161
159
the stretching vibrations of the C@O is poorly expressed. The dis-
tinct decrease of this band intensity was observed while deproto-
nation of neutral form in solution occurs under the action of
NaOH to the DMSO solution.
H2R+
50
40
30
20
10
0
In the 13C NMR spectrum of the sodium salt of fluorescein
methyl ether, one carbon signal was not observed. This missing
signal was assigned to the central carbon atom by HMBC and
HSQC experiments and is due to the equilibrium quinonoid –
lactone. The last-named structure may be either neutral, HL,
which can be formed as a result of incomplete deprotonation
of the dye or hydrolysis by residual water in DMSO-d6, or anio-
nic, Lꢁ.
pH
pH
R–
5.37
5.27
4.90
4.75
4.55
4.52
4.36
3.96
3.64
1.87
2.06
2.26
2.49
2.66
2.87
3.10
3.50
After adding NaOH to the solution of neutral form in DMSO, the
signal d = 153.33 ppm was registered that indicates complete con-
version to the anionic form. The value of chemical shift agrees well
with the value for the central carbon signal of methyl ether ester
(d = 150.14 ppm). Thus, the NMR data do not give evidence of exis-
tence of the Lꢁ species for fluorescein methyl ether.
The addition of trifluoroacetic acid to the salt solution leads to
restoring of the neutral lactonic structure and appearing of signal
83.09 ppm of sp3 hybridization of the central atom.
HR
400
450
λ/nm
500
Fig. 7. Absorption spectra of fluorescein methyl ether aqueous solutions at different
pH and limiting dye forms (heavy lines).
Note, that the presence of excess of NaOH results in appear-
ance of signal 72.63 ppm that is due to the hydroxylation of cen-
tral atom with possible destruction (rupture) of the xanthene
moiety.
Hence, the existence of lactoid monoanion of methyl fluorescein
in solution seems to be unlikely. The same may be deduced for the
parent compound fluorescein; such conclusion is helpful in under-
standing the equilibrium of the latter dye (Scheme 1), because the
yield of the species HRꢁ of fluorescein in solution is relatively
small, especially in organic solvents, and thus its UV–vis and in
particular IR and 13C NMR spectra are less available.
In the case of ethers of eosin and rose Bengal B, the fraction of
the lactonic monoanion is very high [37]. Such difference between
these dyes and the methyl ether of the unsubstituted fluorescein
should be easily explained by the dramatic strengthening of the
acidity of the hydroxy group due to introduction of the two Br or
I atoms in the ortho-position.
transformation of the dye into the anion was reached by introduc-
ing appropriate amounts of diluted NaOH. In non-aqueous solu-
tions, DBU with concentration circa 0.02 mol Lꢁ1 was used as a
deprotonating agent.
The spectra of the anion exhibit strong variation of the shape
and still more of intensity. It should be noted that the effects are
less marked in the hydrogen bond donor (HBD) solvents as com-
pared with that in the non-HBD ones. This may be connected with
the character of solvation of the COOꢁ group. On the other hand,
the variations are much less expressed for the neutral quinonoid
form of the ether-ester, which is unable to lactonization (Fig. 9).
Therefore, we had to clarify if these alterations in molar absorptiv-
ity are caused either by entire solvatochromic effects of the colored
species Qꢁ or by formation of the colorless lactone Lꢁ.
Therefore, the study was furthered with the help of IR and NMR
spectroscopy.
Microscopic dissociation constants
In the IR spectra of sodium salt of fluorescein methyl ether in
solid state as well as in DMSO solution the band corresponding
From the Schemes 1 and 2, the following equations may be
derived:
pKa0 ¼ pk0;OH ꢁ logð1 þ KT þ K=TÞ
ð3Þ
9
8
7
6
5
4
3
2
1
0
0.4
0.3
0.2
0.1
0
pKa1 ¼ pk1;COOH þ logð1 þ KT þ K=TÞ
ð4Þ
1
The difference between fluorescein and its methyl ether con-
sists in the possibility of the parent compound to dissociate in an
additional step (the pKa2 value). Also, it possesses two hydroxy
groups in the xanthene portion, contrary to the sole OH group of
the ether.
Knowing the fractions of the tautomers HQ and HZ (see above),
one can estimate the microscopic ionization constants. In Table 4,
they are compared with those of fluorescein. The similarity of the
corresponding pk values of fluorescein and its methyl ether is strik-
ing. In the case of pk1;COOH and even pk
the absence of conjugation between the xanthene moiety and the
phthalic acid residue.
For the pk1;Z and pk0;OH values, however, it is an insubstantial
coincidence. Indeed, the existence of two OH groups in fluorescein,
fluorescein ethyl ester and 6-hydroxy-9-phenyl fluorone means
that the last-mentioned pk must be log 2 = 0.301 higher if ascribed
to a single OH group. But the pk0;OH values of sole OH group of the
methyl ether ester (2.28) and methyl ether (2.31) are close to the
2
it is natural because of
ꢃ;COOH
400
420
440
460
480
500
520
λ/nm
Fig. 8. Absorption spectra of neutral form of fluorescein methyl ether in aqueous
(dashed line) and 50 mass% aqueous ethanol (solid line) solutions.

160
N.O. Mchedlov-Petrossyan et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 150 (2015) 151–161
30
25
20
15
10
5
30
25
20
15
10
5
1
6
5
2
3
4
5
6
7
3
7
0
0
λ
/nm
520
λ
/nm
520
400
420
440
460
480
500
400
420
440
460
480
500
(a)
(b)
Fig. 9. Absorption spectra of fluorescein methyl ether monoanion (a) and fluorescein methyl ether ester (b) in water (1), methanol (2), 95.6% aqueous ethanol (3), 1-butanol
(4), dichloromethane (5), DMSO (6) acetone (7). Solutions 1–6 contain 1 vol% acetone.
Table 4
sulfonefluorescein, however, predominates in any solvent, owing
to the weak basicity of the SO3ꢁ group; recently it was confirmed
The indices of the microscopic ionization constants of methyl ether of fluorescein and
those of the parent compound.
by quantum-chemical calculations [31].
Water
50 mass% aqueous ethanol
Fluorescein Methyl ether of Fluorescein
pk
Moreover, one may estimate the pk
¼ 4:49 value of fluo-
ꢃ;COOH
rescein methyl ether on aqueous ethanol by using the above
log K=T value and pk0;OH ¼ 2:31.
Methyl ether of
fluorescein
fluorescein
3.48
3.78
3.19
2.89
3.49
3.79
3.10
2.80
5.40
–
2.31
–
5.38
–
2.41
–
pk1;COOH
pk1;Z
pk0;OH
The pKa2 values of fluorescein (7.66) and sulfonefluorescein
(7.63) coincide. As the lactone-like (sultone) tautomer of the HRꢁ
species of the latter dye is improbable, the same may be deduced
for the fluorescein monoanion as well. The difference dpKa
between the pKa2 of these dyes and the average pKa1 value of ethyl
ester of fluorescein and 6-hydroxy-9-phenyl fluorone (which are
also rather close: 6.71 and 6.80) is 0.89, while in water dpKa equals
to 0.48. These differences should be explained in terms of the
Bjerrum – Kirkwood – Westheimer equation [28], which describes
the influence of the charged group on the pKa of the ionizing one.
pk
ꢃ;COOH
values of 6-hydroxy-9-phenyl fluorone (2.34) and fluorescein
(2.41); only for the ethyl ester of fluorescein the pk0;OH value 2.05
is really lower. Hence, it may be concluded that the substitution
of the OH group in the xanthene moiety by the OCH3 group some-
what increases the acidity of the remaining hydroxyl group. In
methyl alcohol [56], this effect is even more pronounced: the
pk0;OH value of fluorescein ethyl ether ester was found to be
4.79 0.03, whereas for fluorescein ethyl ester and
e2NA
dpKa
¼
:
ð6Þ
4
p
ꢅ 8:854 ꢅ 10ꢁ12 ꢅ 2:303RTrDeff
Here e is the elemental electrical charge, NA is the Avogadro
number, R is the gas constant, T is the absolute temperature, r is
the distance between the ionizing and charged groups, and Deff is
the effective permittivity of the space between the two above
groups.
6-hydroxy-9-phenyl
fluorone:
pk0;OH ¼ 5:23 ꢃ 0:10
and
5.14 0.03, respectively. The low value of 4.8 for the unsubstituted
fluorescein may be explained by the contributory uncertainty of
the estimation of the small fraction of the quinonoid tautomer
H2Q absolute methanol [65].
In order to estimate the fraction of the HZ tautomer in aqueous
ethanol, the data for the dye sulfonefluorescein should be used.
Indeed, its pKa1 value is in fact pk1;Z. On the other hand, from
Scheme 2 the following equation may be derived:
It should be noted that somewhat smaller difference following
from the data by Fompeydie and Levillain in 50 vol% aqueous
methanol
(fluorescein:
pKa2 ¼ 6:86,
ethyl
fluorescein:
pKa1 ¼ 6:36) [25] may be explained by the higher relative permit-
tivity value of the solvent as compared with our ethanol–water
mixture. Also, there are no indications of the thermodynamic char-
acter of their pKa s; if they are expressed in the concentration scale,
the re-calculation to the thermodynamic values will increase the
discussed difference.
log K=T ¼ pk0;OH ꢁ pk
¼ pk1;Z ꢁ pk1;COOH
ð5Þ
,
ꢃ;COOH
Using the pKa1 value of sulfonefluorescein, which is in fact pk1;Z
and the pk1;COOH value of fluorescein methyl ether, one can obtain
log K=T ¼ ꢁ2:18. It means that the fraction of the HZ tautomer is
150 times lower than that of the quinonoid HQ, which is in turn
only as low as 0.01. Such dramatic destabilization of the dipolar
tautomer is in line with behavior of the parent compound fluores-
cein [6,7,27–29,58,59,64,65]. The zwitterionic tautomer of
The same equation explains the difference between the
pk1;Z ¼ 3:22 value of sulfonefluorescein in 50% ethanol and the five
pk0;OH of fluorescein, its ethyl ester, methyl ether, methyl
ether-ester, and 6-hydroxy-9-phenylfluorone (2.05–2.41) in the
same solvent. Here, dpKa is within the range of 0.81–1.17.

N.O. Mchedlov-Petrossyan et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 150 (2015) 151–161
161
[10] H. Yao, J.D. Steill, J. Oomens, R.A. Jockusch, J. Phys. Chem. A 115 (2011) 9739.
[11] P.D. McQueen, S. Sagoo, H. Yao, R.A. Jockusch, Angew. Chemie Int. Ed. 49
(2010) 9193.
[12] H. Yao, R.A. Jockusch, J. Phys. Chem. A 117 (2013) 1351.
[13] T. Tanabe, M. Saito, K. Noda, J. Phys. Conf. Ser. 388 (2012) 022106.
[14] T. Tanabe, M. Saito, K. Noda, E.B. Starikov, Eur. Phys. J. D 66 (2012) 163.
[15] T. Ueno, Y. Urano, K. Setsukinai, H. Takakusa, H. Kojima, K. Kikuchi, K. Ohkubo,
S. Fukuzumi, T. Nagano, J. Am. Chem. Soc. 126 (2004) 14079.
[16] Y. Urano, M. Kamiya, K. Kanda, T. Ueno, K. Hirose, T. Nagano, J. Am. Chem. Soc.
127 (2005) 4888.
[17] L.D. Lavis, R.T. Raines, ACS Chem. Biol. 3 (2008) 142.
[18] A. Persat, M.E. Suss, J.G. Santiago, Lab Chip 9 (2009) 2454.
[19] L. Lindqvist, Ark. Kemi 16 (1960) 79.
[20] N.O. Mchedlov-Petrossyan, Zhurn. Anal. Khim. 34 (1979) 1055.
[21] Z. Tamura, T. Morioka, M. Maeda, A. Tsuji, Bunseki Kagaku 43 (1994) 339.
[22] N. Klonis, W.H. Sawyer, J. Fluoresc. 6 (1996) 147.
[23] R. Sjöback, J. Nygren, M. Kubista, Spectrochim. Acta A 51 (1995) L7.
[24] M. Shibata, M. Nakamizo, H. Kakiyama, Nippon Kagaku Kaishi 1972 (1972)
681.
[25] D. Fompeydie, F. Onur, P. Levillain, Bull. Soc. Chim. Fr. 9–10 (1979) 375.
[26] D. Fompeydie, P. Levillain, Bull. Soc. Chim. Fr. 11–12 (1980) 459.
[27] N.O. Mchedlov-Petrossyan, R.S. Mayorga, J. Chem. Soc. Faraday Trans. 88
(1992) 3025.
[28] N.O. Mchedlov-Petrossyan, V.I. Kukhtik, V.D. Bezugliy, J. Phys. Org. Chem. 16
(2003) 380.
[29] N.O. Mchedlov-Petrossyan, N.A. Vodolazkaya, N.V. Salamanova, A.D. Roshal,
D.Y. Filatov, Chem. Lett. 39 (2010) 30.
Conclusions
The spectroscopic study of the protolytic equilibria in solutions
of methyl ether and methyl ether-ester of fluorescein, as well as
related compounds sulfonefluorescein, ethyl ester of fluorescein,
and 6-hydroxy-9-phenylfluorone enable better understanding of
the acid-base dissociation and tautomerism of the widely used
unique parent dye fluorescein. The neutral (molecular) form of flu-
orescein methyl ether in solution is an equilibrium mixture of
three tautomers: zwitter-ionic, HZ , quinonoid, HQ, and colorless
lactonic one, HL. In water, the fractions are correspondingly 11%,
6%, and 83%. On going from water to 50% aqueous ethanol, the frac-
tion of the dipolar tautomer drops sharply to ca. 0.01%, that of the
quinonoid decreases markedly (1%), whereas the colorless lactone
predominates. The last-named tautomer was liberated and charac-
terized by the X-ray analysis that revealed a substantially longer
C–O bond for the central carbon atom as compared with fluores-
cein and common c-lactones.
Another problem that has been considered is the possibility of
formation of monoanion having lactonic structure HL. The utiliza-
tion of UV–visible, IR, and 13C NMR spectroscopy allows to consider
the anion–lactone of fluorescein methyl ether as unlikely. The
same may be deduced for the parent compound fluorescein; such
conclusion is helpful in understanding the equilibrium of the latter
dye (Scheme 1), because the yield of the species HRꢁ of fluorescein
in solution is relatively small, especially in organic solvents, and
thus its UV–vis and in particular IR and 13C NMR spectra are less
available.
[30] N.O. Mchedlov-Petrossyan, N.A. Vodolazkaya, Y.N. Surov, D.V. Samoylov,
Spectrochim. Acta A 61 (2005) 2747.
[31] A.V. Lebed, A.V. Biryukov, N.O. Mchedlov-Petrossyan, Chem. Heterocycl.
Compd. 50 (2014) 336.
[32] S.-C. Chen, H. Nakamura, Z. Tamura, Chem. Pharm. Bull. 27 (1979) 475.
[33] Z.-G. Zhao, T. Shen, H. Xu, Spectrochim. Acta A 45 (1989) 1113.
[34] H. Zhang, M. Zhang, T. Shen, J. Photochem. Photobiol. A 103 (1997) 63.
[35] A. Song, J. Zhang, M. Zhang, T. Shen, J. Tang, Colloids Surfaces A 167 (2000) 253.
[36] X.-F. Zhang, Q. Liu, H. Wang, Z. Fu, F. Zhang, J. Photochem. Photobiol. A 200
(2008) 307.
The microscopic dissociation constants of fluorescein methyl
ether in water and in 50% ethanol coincide with the corresponding
quantities for fluorescein, thus confirming the adequacy of the gen-
eral scheme of protolytic equilibria in solution.
[37] F. Amat-Guerri, M.M.C. López-González, R. Sastre, R. Martínez-Utrilla, Dyes
Pigm. 13 (1990) 219.
[38] Y. Nagase, T. Ohno, T.J. Goto, Pharm. Soc. Japan 75 (1953) 1039.
[39] J.-H. Scharf, Mikroskopie 11 (1956) 261.
[40] V. Zanker, W. Peter, Chem. Ber. 91 (1958) 572.
[41] M.M. Martin, L. Lindqvist, J. Lumin. 10 (1975) 381.
[42] H. Diehl, N. Horchak-Morris, A.J. Hefley, L.F. Munson, R. Markuszewski, Talanta
33 (1986) 901.
Acknowledgements
[43] O. Fischer, E. Hepp, Berichte der Dtsch. Chem. Gesellschaft 46 (1913) 1951.
[44] W.R. Orndorff, R.S. Vose, J. Am. Chem. Soc. 46 (1924) 1836.
[45] R.G. Bates, Determination of pH, Khimiya, Leningrad, 1972.
[46] E. Kozáková, B. Cséfalvayová, J. Zajacová, Chem. Zvesti 36 (1982) 345.
[47] O.V. Dolomanov, L.J. Bourhis, R.J. Gildea, J.A.K. Howard, H. Puschmann, J. Appl.
Crystallogr. 42 (2009) 339.
The authors express their gratitude to Dr. D. Yu. Filatov and Dr.
D. S. Sofronov for measuring the IR spectra and to Dr. A. D. Roshal
for fluorescence measurements.
[48] G.M. Sheldrick, Acta Crystallogr. A 64 (2008) 112.
[49] http://www-chemo.univer.kharkov.ua/kholin/clinp.html.
[50] R.S. Osborn, D. Rogers, Acta Crystallogr. B 31 (1975) 359.
[51] I.N. Polyakova, Z.A. Starikova, B.V. Parusnikov, I.A. Krasavin, G.M. Dobryakova,
B.V. Zhadanov, J. Struct. Chem. 25 (1985) 752.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.saa.2015.05.037.
ˇ
[52] D.-K. Bucar, S. Filip, M. Arhangelskis, G.O. Lloyd, W. Jones, CrystEngComm 15
(2013) 6289.
[53] F.H. Allen, Acta Crystallogr. A 58 (2002) 380.
References
[54] H.-B. Burgi, J.D. Dunitz, Structure Correlation, VCH, Weinheim, 1994.
[55] U. Anthoni, C. Christophersen, P.H. Nielsen, A. Püschl, K. Schaumburg, Struct.
Chem. 6 (1995) 161.
[56] Y.V. Zefirov, Crystallogr. Rep. 42 (1997) 865.
[57] M. Davies, R.L. Jones, J. Chem. Soc. (1954) 120.
[58] N.O. Mchedlov-Petrossyan, R.S. Mayorga, Y.N. Surov, Zhurn. Obsh. Khim. 61
(1991) 225.
[59] N.O. Mchedlov-Petrossyan, Y.N. Surov, S.I. Egorova, R.S. Mayorga, E. Arias
Kordova, Dokl. Akad. Nauk SSSR 317 (1991) 152.
[1] X. Li, X. Gao, W. Shi, H. Ma, Chem. Rev. 114 (2014) 590.
[2] D.S. Pellosi, B.M. Estevao, C.F. Freitas, T.M. Tsbone, W. Caetano, N. Hioka, Dyes
Pigm. 99 (2013) 705.
[3] A. Niazi, A. Yazdanipour, J. Ghasemi, M. Kubista, A. Sheyda, M. Alikhah, Croat.
Chem. Acta 82 (2009) 753.
[4] V.R. Batistela, J. da Costa Cedran, H.P. Moisés de Oliveira, I.S. Scarminio, L.
Ueno, A.E. da Hora Machado, N. Hioka, Dyes Pigm. 86 (2010) 15.
[5] X.-F. Zhang, Photochem. Photobiol. Sci. 9 (2010) 1261.
[6] L.N. Bogdanova, N.O. Mchedlov-Petrossyan, N.A. Vodolazkaya, A.V. Lebed,
Carbohydr. Res. 345 (2010) 1882.
[7] N.O. Mchedlov-Petrossyan, N.A. Vodolazkaya, Y.A. Gurina, W.-C. Sun, K.R. Gee,
J. Phys. Chem. B 114 (2010) 4551.
[8] V.R. Batistela, D.S. Pellosi, F.D. de Souza, W.F. da Costa, S.M. de Olivera Santin,
V.R. de Souza, W. Caetano, H.P. Moisés de Oliveira, I.S. Scarminio, N. Hioka,
Spectrochim. Acta A 79 (2011) 889.
[60] SDBSWeb: http://sdbs.db.aist.go.jp.
[61] N.O. Mchedlov-Petrossyan, Dokl. Akad. Nauk SSSR 293 (1987) 1178.
[62] S. Niizuma, K. Kikuchi, H. Kokubun, J. Photochem. 36 (1987) 51.
[63] B.Y. Frolov, M.A. Babashov, S.G. Kudryavtsev, Zhurn. Fiz. Khim. 66 (1992) 2970.
[64] N.O. Mchedlov-Petrossyan, Kharkov Univ. Bull. 626 (2004) 221.
[65] N.O. Mchedlov-Petrossyan, L.V. Vasetskaya, Zhurn. Obsh. Khim. 59 (1989) 691.
[66] N.O. Mchedlov-Petrossyan, V.V. Ivanov, Zhurn. Fiz. Khim. 81 (2007) 117.
[9] N.G. Greeneltch, A.S. Davis, N.A. Valley, F. Casadio, G.C. Schatz, R.P. Van Duyne,
N.C. Shah, J. Phys. Chem. A 116 (2012) 11863.