Synthesis and chemical characterization of the novel agronomically relevant pentadentate chelate 2-(2-((2-hydroxybenzyl)amino)ethylamino)2-(2- hydroxyphenyl)acetic Acid (DCHA)

Article abstract of DOI:10.1021/jf100994s

Iron chelates analogous to o,o-EDDHA/Fe^3- are the fertilizers chosen to treat iron chlorosis in plants growing on calcareous soil. The isomer o,p-EDDHA/Fe³⁺ presents less stability but faster assimilation by the plant than o,o-EDDHA/

Full text of DOI:10.1021/jf100994s

7908 J. Agric. Food Chem. 2010, 58, 7908–7914
DOI:10.1021/jf100994s
Synthesis and Chemical Characterization of the
Novel Agronomically Relevant Pentadentate Chelate
2-(2-((2-Hydroxybenzyl)amino)ethylamino)-
2-(2-hydroxyphenyl)acetic Acid (DCHA)
†
´
ERNANDEZ
S
ANDRA LO´ PEZ-RAYO, DIANA
Departamento de Quımica Agrı
H
,
AND
J
UAN J. LUCENA
*
´
´
cola, Facultad de Ciencias, Universidad Auto
´
noma de Madrid,
28049 Madrid, Spain. †Present address: Centro de Ciencias Medioambientales, Consejo Superior de
Investigaciones Cientı
´
ficas, Serrano 115, bis, E-28006 Madrid, Spain
AND
IGUEL A. SIERRA
mica, Universidad Complutense,
R
OSA
E
SCUDERO, MAR GO´ MEZ-GALLEGO
,
M
*
Departamento de Quı
´
mica Orga
´
nica, Facultad de Quı
´
28040 Madrid, Spain
Iron chelates analogous to o,o-EDDHA/Fe3þ are the fertilizers chosen to treat iron chlorosis in plants
growing on calcareous soil. The isomer o,p-EDDHA/Fe3þ presents less stability but faster assimila-
tion by the plant than o,o-EDDHA/Fe3þ, because only five coordinating groups are able to complex
Fe3þ. The new chelating agent 2-(2-((2-hydroxybenzyl)amino)ethylamino)-2-(2-hydroxyphenyl)acetic
acid (DCHA) has been synthesized to obtain an iron fertilizer with intermediate stability between
o,o-EDDHA/Fe3þ and o,p-EDDHA/Fe3þ and with fast assimilation. Its synthesis has been done
starting from phenol, N-acetylethylendiamine, glyoxylic acid, and NaOH in a three-step sequence.
The purity of the DCHA chelating agent, its protonation, and Ca2þ, Mg2þ, Fe3þ, and Cu2þ stability
constants, together with its ability to maintain iron in solution in different agronomic conditions, have
been determined. The results indicate that the chelate DCHA/Fe3þ has intermediate stability between
those of o,o-EDDHA/Fe3þ and o,p-EDDHA/Fe3þ complexes and that it is capable of maintaining the
Fe3þ in agronomic conditions. This new chelating agent may be effective in correcting iron chlorosis in
plants.
KEYWORDS: Iron chelates; iron chlorosis; fertilizers; o,o-EDDHA/Fe3þ; o,p-EDDHA/Fe3þ; 2-(2-((2-hydroxy-
benzyl)amino)ethylamino)-2-(2-hydroxyphenyl)acetic acid (DCHA)
INTRODUCTION
p,p-EDDHA, all of them are able to bind iron, but only
o,o-EDDHA and o,p-EDDHA have been proved to be effective
as iron suppliers to plants (7, 10). The chelate o,o-EDDHA/Fe3þ
is highly stable (K ≈ 1035) (11), whereas o,p-EDDHA/Fe3þ
presents a significantly lower stability (K ≈ 1029) (12) than
o,o-EDDHA/Fe3þ due to the pentacoordinated nature of the
ligand that requires the incorporation of a water molecule to fill
the octahedral environment around the metal. Despite this lower
stability, o,p-EDDHA/Fe3þ is a better substrate for the iron
chelate reductase of Fe-stressed cucumber plants (10) than
o,o-EDDHA/Fe3þ. It was proposed that the open nature of the
o,p-EDDHA/Fe3þ makes the iron more accessible to the iron che-
late reductase before the reduction step. Hence, o,p-EDDHA/Fe3þ
is a faster substrate for plant iron assimilation than o,o-EDDHA/
Fe3þ. Recent studies found that Fe-stressed soybean plants show
a faster response too,p-EDDHA/Fe3þ than to o,o-EDDHA/Fe3þ
in hydroponics (10). The counterpart to this higher activity is the
lower stability of the complex that reduces the available amount
of o,p-EDDHA/Fe3þ in the alkaline soil solutions compared to
the more stable o,o-EDDHA/Fe3þ (11). When these chelates are
Iron chlorosis is a major problem for crops cultivated in
alkaline and calcareous soils (1), and the use of iron chelates
derived from o,o-EDDHA/Fe3þ (Figure 1) in soil applications is
the standard treatment to alleviate this plant disease (2, 3).
Although synthetic aminopolycarboxylate chelating agents are
now under scrutiny due to their influence on metal availability
and mobility, especially due to their persistence in the environ-
ment (4,5), nowadays they are the most efficient soil fertilizers to
treat iron chlorosis in plants growing in calcareous soil. Com-
mercial EDDHA/Fe3þ fertilizers are made by incorporating an
iron salt to a mixture of reaction products obtained in the
industrial synthesis production of o,o-EDDHA. This mixture
contains o,o-EDDHA (1) as the major component, together with
the regiosiomers o,p-EDDHA (2) and p,p-EDDHA in variable
amounts, and is accompanied by polycondensation products
and other byproducts recently revealed (6-9). Except for
*Authors to whom correspondence should be addressed [(J.J.L.)
e-mail juanjose.lucena@uam.es; (M.A.S.) sierraor@quim.ucm.es].
pubs.acs.org/JAFC
Published on Web 06/09/2010
© 2010 American Chemical Society

Article
J. Agric. Food Chem., Vol. 58, No. 13, 2010 7909
Figure 1. Structures of chelating agents o,o-EDDHA (1), o,p-EDDHA (2), and DCHA (3) and their Fe(III) chelates.
applied at once in calcareous soils, o,o-EDDHA/Fe3þ pro-
vides better iron nutrition to strategy I plants than o,p-EDDHA/
Fe3þ (13, 14).
δ 176.8 (CdO), 174.9 (CdO), 156.6, 130,4, 130.3, 122.8, 120.1, 116.8
(ArC), 64.9 (CH), 45.9 (CH2), 38.3 (CH2), 22.2 (CH3).; IR (KBr) ν 3423
(NH), 1656 (CdO, amide I), 1578, 1444 (NCdO, amide II), 1377, 1104,
742 cm-1
.
These findings led us to design a novel pentacoordinate Fe
chelate (DCHA/Fe3þ) having two phenolate groups and a car-
boxylate moiety, in addition to the two amino groups, coordi-
nated to iron. This compound will be essentially different from
o,p-EDDHA/Fe3þ, and it is predicted to combine a good stability
in nutrient solutions and calcareous soils (due to the presence of
the two phenolates as in o,o-EDDHA) and a fast action to solve
iron chlorosis due to its open nature like that of o,p-EDDHA/
Fe3þ. The new chelate DCHA/Fe3þ has been synthesized and
then studied using a previously described methodology (11) to
predict its capacity to solve iron chlorosis. This methodology
includes the calculation of the protonation and stability constants
of DCHA with Fe3þ and other cations existing in soils such as
Ca2þ, Mg2þ, and Cu2þ and the modeling of the behavior of
DCHA/Fe3þ under agronomic conditions.
The hydrolysis of the acetamido group in 4 was carried out by reflux
with HCl 15% v/v (60 mL, 2 h) to yield 2.83 g (90%) of amine
hydrochloride 5 as a yellow solid after evaporation of the solvent: 1H
NMR (D2O) δ 7.29-7.23 (m, 2H, ArH), 6.89-6.86 (m, 2H, ArH), 5.08 (s,
1H, CH), 3.28-3.20 (m, 4H, 2CH2); 13C NMR (D2O) δ 171.0 (CdO),
155.14, 132.8, 131.8, 121.2, 116.8, 116.3 (ArC), 61.2 (CH), 43.1 (CH2),
35.7 (CH2); IR (KBr) ν 3000-2400 (broad, NH2þ), 1728 (CdO), 1230,
761 cm-1
.
Synthesis of Imine 6. Amine hydrochloride 5 (300 mg, 1.06 mmol) was
dissolved in 3 mL of water and the pH of the solution adjusted to 7.2 with
10% NaHCO3. Then, a solution of salicylaldehyde (130 mg, 1.06 mmol) in
1 mL of absolute EtOH was added. The mixture was stirred at room
temperature for 1 h. The yellow precipitate formed after 1 h at room
temperature was filtered and dried, yielding 211 mg (63%) of imine 6 as
yellow solid: 1H NMR (DMSO-d6) δ 8.54 (s, 1H, CHdN), 7.48 (d, J =
7.5 Hz, 1H, ArH), 7.34 (t, J = 7.62 Hz, 1H, ArH), 7.30 (d, J = 7.32 Hz,
1H, ArH), 7.18 (t, J = 7.32 Hz, 1H, ArH), 6.92-6.70 (m, 4H, ArH), 4.62
(s, 1H, CH), 3.83 (m, 2H, CH2), 3.15 (m, 1H, CH2), 3.02 (m, 1H, CH2); 13
C
EXPERIMENTAL PROCEDURES
NMR (DMSO-d6) δ 169.9 (CdO), 167.6 (CHdN), 160.3, 156.2, 136.4,
132.5, 131.8, 128.8, 127.5, 122.4, 118.8, 118.6, 116.6, 116.5 (ArC), 61.02
(CH), 55.61 (CH2), 46.74 (CH2); IR (KBr) ν 3430-2934 (broad), 1665
(CdO), 1623, 1452 (CdN), 1384, 834, 760, 700 cm-1; ESI-MS m/z 337
[M þ Na]þ; HRMS (ESI) calcd for C17H19N2O4 ([M þ H]þ), 315.1360;
found, 315.1354.
General. All reagents used in this work were of analytical grade. All
aqueous solutions were prepared with water according to type I grade (15).
NMR spectra were recorded at 22 °C on a Bruker Avance DPX 300 (300
MHz for 1H, 75 MHz for 13C) or on a Bruker Avance 500 (500 MHz for
1H, 125 MHz for 13C). Chemical shifts are given in part per million relative
to D2O (1H, 4.78 ppm), D2O/Na2CO3 (13C, 165.7 ppm), or DMSO-d6 (1H,
2.5 ppm; 13C, 40.6 ppm). IR spectra were taken on a Perkin-Elmer 781
spectrometer. ESI-MS spectra were carried out in MeOH, using an
ESQUIRE-LC (Bruker Daltonic, Bremen, Germany) ion trap spectro-
meter in negative mode of detection. The stainless steel capillary was held
at a potential of 5.0 kV. Nitrogen was used as nebulizer gas at a flow rate of
3.98 L min-1 (nebulizer pressure = 11 psi) at 150 °C. All commercially
available products were used without further purification. The details of
the methods employed to determine the purity, the stability constants, the
pM values, and the stability of the chelating agents have been described in
an earlier paper (11).
Synthesis of DCHA. Synthesis of the Amine Hydrochloride 5.
Monoacetyl ethylenediamine (1.08 g, 10.62 mmol) was added to a stirred
flask containing 25.0 g (265.6 mmol) of melted phenol and 0.64 g (7.97
mmol) of NaOH (50% w/w) at 30-35 °C. Then, 1.57 g (10.62 mmol) of
glyoxylic acid (50% in water) was added slowly dropwise to the solution,
keeping the temperature below 40 °C. The mixture was kept at 70-75 °C
for 2 h and then at room temperature for 20 min. After the addition of
water (60 mL), the reaction was extracted with CH2Cl2 (3 ꢀ 20 mL), and
the solvent was evaporated under reduced pressure to yield 1.9 g (96%) of
amide 4 as a solid: 1H NMR (D2O) δ 7.17-7.04 (m, 2H, ArH), 6.82-6.66
(m, 2H, ArH), 4.07 (s, 1H, CH), 3.26-3.16 (m, 2H, CH2), 2.84-2.74 (m,
1H, CH2), 2.71-2.54 (m, 1H, CH2), 1.83 (s, 3H, CH3); 13C NMR (D2O)
Synthesis of DCHA (3). A suspension of imine 6 (250 mg, 0.79 mmol)
and 10% catalyst Pd(C) (5% w/w) in 30 mL of MeOH was stirred under
H2 pressure (40 psi) for 6 h at room temperature. The catalyst was removed
by filtration on Celite and the solvent by distillation under reduced
pressure. Amino acid 3 was obtained as a beige solid (240 mg, 95%): 1H
NMR (D2O) δ 7.20-7.01 (m, 4H, ArH), 6.79-6.66 (m, 4H, ArH), 4.36
(s, 1H, CH), 3.87 (s, 2H, CH2), 2.84-2.69 (m, 4H, CH2); 13C NMR (D2O,
CD3OD) δ 176.9 (CdO), 157.3, 157.1, 132.0, 131.0, 130.3, 129.8, 129.6,
124.9, 121.0, 119.7, 118.0, 117.3 (ArC), 64.8 (CH), 48.9, (CH2), 45.9 (CH2),
44.3 (CH2); IR (KBr) ν 3367-3057 (broad), 1616 (CdO), 1589, 1558,
1460, 1394, 1274, 756 cm-1; ESI-MS m/z 339 [M þ Na]þ; ESI-MS
m/z 317.1 [M þ H]þ; HRMS (ESI) calcd for C17H21N2O4 ([M þ H]þ),
317.1496; found, 317.1495.
Determination of the Purity of the Chelating Agent. The ligand in
concentration 1 ꢀ 10-4 M was dissolved in NaOH calculated to be 3 times
the molaramount of theligand, andthepH was fixedat 6 by the addition of
2 mM MES buffer (2-(N-morpholino)ethanesulfonic acid). Ionic strength
was adjusted to 0.1 M by the addition of NaCl. The experimental solution
was photometrically titrated at 25.0 ( 0.5 °C during the addition of 4.58 ꢀ
10-4 M Fe3þ standard solution until the absorbance at 480 nm presented
no changes. Finally, the purity of the chelating agent was calculated
following the mathematical procedure previously described (11).

´
Lopez-Rayo et al.
7910 J. Agric. Food Chem., Vol. 58, No. 13, 2010
Determination of Stability Constants by Potentiometric Data.
H
Table 1. Composition of Theoretical Soil Model with Unlimited Cu2þ Avail-
The lowest protonation constants (K3 and K4H), corresponding to the
ability Used To Predict the Stability of DCHA/Fe3þ in Soil Conditions
two amino groups, were determined by potentiometric titration (16)
of DCHA chelating agent solution (1 ꢀ 10-3 M; μ = 0.1 M (NaCl)),
previously dissolved in NaOH, with 0.0571 M HCl solution at 25 °C under
N2 atmosphere. Ca2þ and Mg2þ solutions were added in 1:1 ligand-to-
metal ratio before the titration to determine the Ca2þ and Mg2þ stability
constants. Total Ca2þ and Mg2þ concentrations were measured by atomic
absorption spectrophotometry. The stability constants were calculated
using the program Hyperquad 2006 (17).
component
equilibrium
log K0
2-
CO2(g) (0.0003 atm)
CO2(g) þ H2O / 2Hþ þ CO3
-18.15
possible solids
soil Ca / Ca2þ
soil Ca
soil Mg
calcite
-2.50
-3.00
-8.41
-3.00
soil Mg / Mg2þ
CaCO3 / Ca2þ þ CO3
2-
2-
dolomite
CaMg(CO3)2 / Ca2þ þ Mg2þ þ 2CO3
Determination of Stability Constants by Spectrophotometric
H
Data. The first two protonation constants (K1 and K2H) of the ligand
were measured spectrophotometrically (18) because the combination of
protons with the phenolic groups is accompanied by extensive changes in
the absorption spectra at 393.8 and 236.8 nm. Twelve DCHA solutions
(1 ꢀ 10-4 M; μ = 0.1 M (NaCl)) were prepared, and the pH was adjusted
from 10.00 to 13.30 with 0.3-0.5 pH intervals. The spectra were recorded
at 200-400 nm in a Shimazdu UV-vis spectrophotometer (Kyoto, Japan)
for all solutions. The wavelength on the maximum absorbance and molar
absorptivities of L4- and LH22- species were initially estimated at pH 12.9
and 9.9, respectively, and used as seeds for the calculations (11). For a
better result, the constant K2H was also refined with the other protonation
constants calculated by the potentiometric method using Hyperquad.
Stability constants for the Fe3þ and Cu2þ chelates were calculated from
spectrophotometric data obtained after titration (11). Solutions of Fe3þ
and Cu2þ chelates (1:1 metal/ligand ratio, μ = 0.1 M (NaCl)) at 1 ꢀ 10-4
and 1 ꢀ 10-3 M, respectively, were prepared under N2 at 25.0 ( 0.5 °C, by
slow addition of Fe3þ or Cu2þ standard solutions. The pH was lowered to
1 by the addition of HCl and titrated with aqueous 0.200 M NaOH titrant
to pH 12. The Fe3þ chelate solution was back-titrated to pH 2.5, too. The
absorbance of the solutions was measured at 430 and 490 nm for the Fe3þ
chelate and at 650 nm for the Cu2þ chelate. Total Fe3þ and Cu2þ
concentrations were measured by atomic absorption spectrophotometry.
infinite solids
soil Cu þ 2Hþ / Cu2þ
soil Fe þ 3Hþ / Fe3þ
soild Zn þ 2Hþ / Zn2þ
soil Cu
soil Fe
soil Zn
2.80
2.70
5.80
conditions, imine 6 precipitated in the reaction medium and was
obtained by filtration in 65% yield. Subsequent reduction of the
imino group with H2/Pt(C) afforded 2 (95% yield), as a pure
compound directly from the reaction medium (Scheme 1). The
use of NaBH4 or NaCNBH3 also produced DCHA in nearly
quantitative yields (spectroscopic). However, in these cases, the
product is heavily contaminated with variable amounts of salts
and its purification could not be achieved. Although the synthesis
of imine 6 has not been reported in the literature, the oxovanad-
ium(IV) complex of 6 (V4þO-EHGS) was described by Pecoraro
and co-workers as a product derived from the oxidative decarbo-
nylation of the monooxovanadium complex V5þO-o,o-EDDHA
(21-23). Transmetalation of V4þO-EHGS with FeCl3 allowed
the synthesis and chemical characterization of the ferric chelate of
6, which was used as a model compound to study the iron binding
in proteins (24). Structurally related manganese complexes were
also reported (25).
H
The protonation constants (K1 and K2H) and the stability constants
(KFeL, KFeHL, KFeoHL, KCuL, KCuHL, and KCuH2L) were calculated from the
data by an in-house program using Microsoft Excel Solver (11).
Fe3þ Chelate Stability versus pH. The behavior of the DCHA/Fe3þ
chelate was studied in agronomic conditions by modeling. The chelate
stability in different conditions was obtained using the equilibrium
speciation model VMINTEQ program (19). The component database
was modified to include the DCHA chelating agent as new component.
The thermodynamic database was also modified including every proton-
ation, Ca2þ, Mg2þ, Cu2þ, and Fe3þ stability constants for each species
formed. Percentage of Fe3þ chelatedby DCHA was calculated over the pH
4-13 range for two agronomic conditions. The values were also calculated
for rac-o,o-EDDHA, meso-o,o-EDDHA, o,p-EDDHA, and EDTA che-
lating agents using the two models. The total concentrations of the ligands
and Fe3þ were 100 μM.
Scheme 1. DCHA Synthetic Sequence
In the first model, Hoagland nutrient solution (20) was used as
theoretical model to study the behavior of DCHA/Fe3þ, using Fe-
(OH)3amp as possible solid controlling the system solubility. The behavior
in soil conditions was calculated in the second model. All soil components
that could have some effect on DCHA/Fe3þ stability were considered.
Two soil types with unlimited and limited Cu2þ availability, respectively,
were proposed to predict the stability of DCHA/Fe3þ with high and low
Cu2þ levels in soil, respectively. The composition of the theoretical soil is
described in Table 1.
RESULTS AND DISCUSSION
The synthesis of (2-(2-((2-hydroxybenzyl)amino)ethylamino)-
2-(2-hydroxyphenyl)acetic acid) (3, DCHA) started from phenol,
N-acetylethylenediamine, glyoxylic acid, and NaOH. After an
extensive study of the reaction conditions, the acetamide 4 was
obtained in nearly quantitative yield as a single ortho isomer. No
traces of the corresponding para isomer were observed (1H NMR
of the crude reaction mixtures). The acetamide 4 was hydrolyzed
with concentrated HCl, and the hydrochloride 5 was obtained
after removal of the solvent under reduced pressure. Condensa-
tion of hydrochloride 5 with salicylaldehyde was carried out in
H2O/EtOH at pH 7.2 (adjusted with 10% NaHCO3). Under these
DCHA/Fe3þ was first characterized to predict its capacity to
solve iron chlorosis. The used methodology has been previously
reported by us (11). This methodology includes the calculation of
the protonation and stability constants of DCHA with Fe3þ and
other cations already present in soils such as Ca2þ, Mg2þ, and
Cu2þ and the modeling of the behavior of DCHA/Fe3þ under
agronomic conditions. The titrimetric purity of the chelating
agent DCHA determined by photometry (with Fe3þ) was

Article
J. Agric. Food Chem., Vol. 58, No. 13, 2010 7911
Table 2. Log Protonation and Log Stability Constantsa for the Chelating Agents (Charges Are Omitted for Simplicity)
quotient
[HL]/[H][L]
DCHA (3)
11.16
rac-o,o-EDDHA (7)
meso-o,o-EDDHA (7)
o,p-EDDHA (16)
11.88
10.80
11.90
10.89
11.18
[H2L]/[H][HL]
[H3L]/[H][H2L]
[H4L]/[H][H3L]
9.33 ( 0.06
7.89 ( 0.13
5.98 ( 0.16
10.18 ( 0.04
8.65 ( 0.05
6.19 ( 0.02
8.67 ( 0.01
6.28 ( 0.11
8.58 ( 0.04
6.16 ( 0.02
[CaL]/[Ca][L]
4.87 ( 0.28
14.51 ( 0.26
23.88 ( 0.17
5.00 ( 0.22
-4.31 ( 0.19
7.99 ( 0.42
17.42 ( 0.39
26.87 ( 0.37
10.13 ( 0.03
7.56 ( 0.49
17.10 ( 0.65
26.41 ( 0.64
9.44 ( 0.08
4.12 ( 0.10
14.27 ( 0.16
23.23 ( 0.32
5.64 ( 0.16
[CaLH]/[Ca][H][L]
[CaLH2]/[Ca] [H]2[L]
[MgL]/[Mg][L]
[MgLOH] [H]/[Mg][L]
[MgLH]/[Mg] [H][L]
[MgLH2]/[Mg][H]2[L]
17.51 ( 0.25
26.56 ( 0.35
15.55 ( 0.03
23.83 ( 0.32
[FeL]/[Fe][L]
27.94 ( 0.12
30.16 ( 0.28
35.86
35.08
34.15
36.56
28.72 ( 0.05
35.02 ( 0.05
37.35 ( 0.10
19.45 ( 0.19
21.74 ( 0.38
30.96 ( 0.09
36.17 ( 0.12
38.14 ( 0.07
[FeHL]/[Fe][H][L]
[FeH2L]/[Fe][H]2[L]
[FeOHL] [H]/[Fe][L]
[CuL]/[Cu][L]
20.18 ( 0.21
22.37 ( 0.18
27.52 ( 0.06
31.88 ( 0.03
23.12
22.81
24.94 ( 0.05
32.87 ( 0.04
37.33 ( 0.07
23.68 ( 0.02
32.30 ( 0.00
37.25 ( 0.01
[CuHL]/[Cu] [H][L]
[CuH2L]/[Cu][H]2[L]
[CuH3L]/[Cu][H]3[L]
a μ = 0.1 M (NaCl); t = 25 °C.
>87% in all cases, which is in good agreement with the estimated
purity obtained by H NMR (considering that the error of this
This behavior is similar to that of o,p-EDDHA/Fe3þ. An iso-
sbestic point is shown at 450 nm, indicating that there are only
two species with different absorptions in the pH range of
7.38-11.83 (in o,p-EDDHA the isosbestic point appears at
470 nm). The optical absorption spectra obtained for DCHA/
Cu2þ present the same tendency as for the Cu2þ complexes of
o,o-EDDHA and o,p-EDDHA (11, 26).
1
technique in our instrumental conditions is (4%). The molar
absorptivity at pH 6.0 of DCHA/Fe3þ (4796 ( 30 L mol-1 cm-1
is similar to that of o,o-EDDHA/Fe3þ (4721 ( 16 L mol-1 cm-1
as was expected considering their structural similarities.
)
)
Stability Constants. The protonation and stability constants of
DCHA are shown in Table 2. These values are compared with
those of o,o-EDDHA, rac-o,o-EDDHA, meso-o,o-EDDHA, and
o,p-EDDHA obtained in earlier works using the same methodo-
logy (11, 26).
DCHA/Cu2þ has stability constants between those of o,o-
EDDHA/Cu2þ and o,p-EDDHA/Cu2þ chelates (Table 2). The o,
o-EDDHA and DCHA ligands are able to complex Cu2þ with
two phenolate and two amine groups. In the case of o,p-EDDHA,
the phenolate group placed in the para position of one of the
aromatic rings cannot bind Cu2þ and a carboxylate forms the
bond instead. The affinity of phenolate groups for Cu2þ is evident
because the stability constant of DCHA/Cu2þ is higher than that
o,p-EDDHA/Cu2þ. The protonated and diprotonated species
of DCHA/Cu2þ (see CuHL, CuH2L in Scheme 2) are similar to
those of o,o-EDDHA/Cu2þ and o,p-EDDHA/Cu2þ (11, 26).
However, the diprotonated species CuH2L produce an important
effect on the Cu2þ stability constant of DCHA. Whereas in
o,o-EDDHAH2/Cu2þ, Cu is bonded by two amino and two
carboxylic groups, in DCHAH2/Cu2þ (CuH2L in Scheme 2) Cu
needs a water molecule to occupy one of the coordination posi-
tions. Similarly, we observed that in CuLH species, the second
carboxylate group should produce a stabilization effect in the
o,o-EDDHA and o,p-EDDHA chelates, possibly due to the for-
mation of the pentadentated chelates. This effect cannot occur in
DCHA/Cu2þ because of the lack of the second carboxylate
group. Then DCHA/Cu2þ species at low pH (protonated and
diprotonated) present lower stability than those of o,o-EDDHA/
Cu2þ and o,p-EDDHA/Cu2þ chelates (Table 2). The main species
present in the pH range typical of calcareous soils (over pH 7.5)
are [CuL]- for DCHA and [CuHL]- for o,o-EDDHA and
o,p-EDDHA.
The protonation constants of DCHA are in general lower than
those of the o,o-EDDHA meso and racemic isomers and those of
o,p-EDDHA. Because the charge of DCHA (L3-) is lower than
those of o,o-EDDHA and o,p-EDDHA (L4-), the acidity of
phenolate and amine groups is higher in DCHA, giving a lower
affinity with protons. The same tendency was observed for the
protonation constants of ethylendiaminetetracetic acid (EDTA)
(L4-; log KH = 10.19 (H1), 6.13 (H2), 2.69 (H3), 2.00 (H4)) versus
HEDTA (L3-; log KH = 9.81 (H1), 5.37 (H2), 2.60 (H3)), where a
hydroxymethyl group unable to bond metals replaces one car-
boxylic group of EDTA (27).
DCHA forms square-planar complexes with Ca2þ and Mg2þ
.
The magnitudes of DCHA/Mg2þ stability constants are higher
than those of DCHA/Ca2þ (Table 2). This tendency is analogous
to that observed for o,o-EDDHA, where the presence of pheno-
late groups favors the binding to small metals such as Mg2þ (11).
The species MgLOH formed for the DCHA ligand stabilized the
chelate at high pH by the deprotonation of one axial molecule of
H2O in the MgL chelate. The low values of the DCHA/Ca2þ and
DCHA/Mg2þ stability constants with respect to those of o,o- and
o,p-EDDHA indicate the poor affinity of these cations with
DCHA. This fact is an advantage for the use of the chelate
DCHA/Fe3þ in calcareous soils, because a lower competence
between Ca2þ and Mg2þ and Fe3þ is expected.
The situation with Fe3þ is clearly different. The values of
stability constants of DCHA/Fe3þ (Table 2) were considerably
lower than those of o,o-EDDHA/Fe3þ and in the range of those
measured for o,p-EDDHA/Fe3þ. The stronger binding of Fe3þ in
o,o-EDDHA derives from its closed octahedrical structure with
six bonds between the ligand and the metallic nucleus. However,
Visible optical absorption spectra for both DCHA/Fe3þ and
DCHA/Cu2þ are shown in Figure 2. Whereas the absorbance
maximum of the o,o-EDDHA/Fe3þ is constant at 480 nm through-
out the pH range (26, 28), the maximum absorbance of DCHA/
Fe3þ reveals a shift from 500 nm at pH 4.16 to 430 nm at pH 11.83.

´
Lopez-Rayo et al.
7912 J. Agric. Food Chem., Vol. 58, No. 13, 2010
Figure 2. Absorption spectra of DCHA/Fe3þ and DCHA/Cu2þ as a function of pH. [DCHA/Fe3þ] = 1 ꢀ 10-4 M, [DCHA/Cu2þ] = 1 ꢀ 10-3 M, t = 25 °C, μ =
0.100 M (NaCl).
Scheme 2. DCHA/Cu2þ Species Formed
the similarity between the values obtained for the stability
constants of the DCHA and o,p-EDDHA ligands is interest-
ing, both having five bonds with the iron nucleus, but two
phenolate-Fe bonds in DCHA against one phenolate-Fe and
one carboxylate-Fe bond in o,p-EDDHA. Both DCHA and
o,p-EDDHA complexes should have a water molecule occupy-
ing the remaining coordination position in the FeL species
(Scheme 3). Below pH 2, one of the phenolates of DCHA/Fe3þ
takes a proton to form FeLH and another water molecule
occupies the vacant position. Over pH 7.8, a proton from the
coordinated water molecule in FeL is released, forming the
FeOHL species (Scheme 3). This behavior is analogous to that
observed in o,p-EDDHA/Fe3þ
.
It is known that the replacement of two carboxylic groups
of EDTA/Fe3þ by two phenolate groups (as in the chelate
o,o-EDDHA/Fe3þ) produces a noticeable increase in stability
(10 units in log K) (29). Therefore, an increment of around 5 units
of log K was expected with the substitution of one of the
carboxylate groups of the complex o,p-EDDHA/Fe3þ by a
phenolate group to form DCHA/Fe3þ. However, the Fe3þ
stability constants of DCHA calculated in this paper are analo-
gous to those obtained for o,p-EDDHA/Fe3þ (11).

Article
J. Agric. Food Chem., Vol. 58, No. 13, 2010 7913
Scheme 3. DCHA/Fe3þ Species Formed
calculated percentage of iron that remains chelated in nutrient
solution (A) and in soil conditions with limited (B) and unlimited
Cu2þ (C) for DCHA/Fe3þ compared to o,p-EDDHA/Fe3þ, rac-
o,o-EDDHA/Fe3þ, meso-o,o-EDDHA/Fe3þ, and EDTA/Fe3þ
.
In nutrient solution the polyphenolic chelates behave similarly,
maintaining the total chelated iron in solution at agronomic pH
values (Figure 3A). In the soil model (Table 1) the main compo-
nents that can affect the iron chelate stability are included.
Calcium and magnesium equilibria are included in the soil model,
because these elements are considered to be the main competitors
of weak chelating agents, such as EDTA (30) in calcareous
conditions. Cu2þ equilibrium has been included in the soil model
to determine the ability of this metal to replace Fe3þ from
DCHA/Fe3þ. The affinity of o,p-EDDHA to Cu2þ produces a
decrease of the iron chelate (26), and the same behavior is
observed for DCHA but to a lesser extent (Figure 3A,B).
Normally, calcareous soils to which iron chelates are frequently
applied also present low Cu2þ availability. The Cu2þ concentra-
tion in model C (near 1 ꢀ 10-4 M at pH 8 for DCHA/Fe3þ) is
high and unusual in soils (26). This range of Cu2þ concentrations
is found only in polluted soils, for example, by the addition of
fungicides (31). Then, in model B the total Cu2þ concentration in
solution is limited to 1 ꢀ 10-5 M. In this case, DCHA/Fe3þ is
stable until pH 11.5 with a behavior similar to that of rac-o,o-
EDDHA/Fe3þ and meso-o,o-EDDHA/Fe3þ and slightly higher
than o,p-EDDHA/Fe3þ. In the three soil models EDTA/Fe3þ is
not effective enough in maintaining Fe3þ in solution at agro-
nomic pH.
On the basis of the theoretical modeling of the chelates, it may
be concluded that DCHA/Fe3þ could be used as a ferric chelate to
correct iron chlorosis, and it may be better than EDTA/Fe3þ and
o,p-EDDHA/Fe3þ for agronomic purposes. This assertion can be
made despite the stability constants that show that the affinity of
DCHA toward Fe3þ is lower than that of o,p-EDDHA. However,
two additional factors improve the stability of the DCHA/Fe3þ in
calcareous soils. First, DCHA forms the FeOHL species at lower
pH values than o,p-EDDHA (7.76 for DCHA and 9.27 for o,p-
EDDHA) and, second, presents lower charge ([FeOHDCHA]-)
than o,p-EDDHA ([FeOH-o,p-EDDHA]2-).
Figure 3. Percentageofchelating agents insolutionthat arebindingFe3þ
:
comparison of the behavior of (1) DCHA, (2) o,p-EDDHA, (3) rac-o,o-
EDDHA, (4) meso-o,o-EDDHA, and (5) EDTA in (A) nutrient solution and
in soil conditions with (B) limited Cu2þ and (C) unlimited Cu2þ
.
Conclusions. The synthesis of (2-(2-((2-hydroxybenzyl)amino)-
ethylamino)-2-(2-hydroxyphenyl)acetic) (DCHA/Fe3þ), a novel
pentacoordinate ferric chelate having two phenol groups and a
carboxylate moiety (in addition to the two amino groups)
coordinated to Fe3þ, has been effected in four steps from phenol,
N-acetylethylendiamine, glyoxylic acid, and NaOH. The purity of
the chelating agent, its protonation, Ca2þ, Mg2þ, Fe3þ, and Cu2þ
stability constants, together with its ability to maintain iron
in solution in different conditions, have been determined. The
results obtained in this work indicate that DCHA/Fe3þ presents
stability constants similar to those of o,p-EDDHA/Fe3þ, but the
theoretical speciation indicates that the species FeOHL favored a
higher stability of DCHA/Fe3þ at calcareous pH. It is predicted
that DCHA/Fe3þ combines a good stability in nutrient solution
and calcareous soils (due to the presence of two phenolate groups
The results above indicated that the increase in stability is due
better to the close geometry of the chelate than to the nature of the
coordinating functionalities, namely, phenolate versus carboxy-
late. The implications of this observation in the mechanism of
action of these complexes (their interaction with the enzyme
FCR) are being explored by experimental and computational
tools in our laboratories.
Fe3þ Chelate Stability versus pH. The simple comparison of
the stability constants of DCHA and the isomers of EDDHA
may not correctly reflect the relative effectiveness of DCHA
in agronomic conditions. It is necessary to know the affinities of
the ferric and competing ions, mainly Cu2þ, for the ligands
at a given pH in agronomic conditions. Figure 3 presents the

´
Lopez-Rayo et al.
7914 J. Agric. Food Chem., Vol. 58, No. 13, 2010
in its structure) and a fast action to relieve iron chlorosis due to its
open nature similar to that of o,p-EDDHA/Fe3þ. Efforts are
underway in our laboratories to determine the agronomical
efficiency and mode of action of this new iron chelate.
Fe(o,o-EDDHA) and Fe(o,p-EDDHA) isomers in supplying Fe to
strategy I plants differs in nutrient solution and calcareous soil.
J. Agric. Food Chem. 2008, 56, 10774–10778.
(14) Schenkeveld, W. D. C.; Dijcker, R.; Reichwein, A. M.; Temminghoff,
E. J. M.; van Riemsdijk, W. H. The effectiveness of soil-applied
FeEDDHA treatments in preventing iron chlorosis in soybean
as a function of the o,o-FeEDDHA content. Plant Soil 2008, 303,
161–176.
(15) ASTM. Standard specification for reagent water. Annual Book of
ASTM Standards; Philadelphia, PA, 1995; 11.01.
(16) Martell, A. E.; Motekaitis, R. J. Determination and Use of Stability
ABBREVIATIONS USED
DCHA, 2-(2-((2-hydroxybenzyl)amino)ethylamino)-2-(2-
hydroxyphenyl)acetic acid; o,o-EDDHA, ethylenediaminedi(2-
hydroxyphenylacetic) acid; o,p-EDDHA, ethylenediamine-N-(o-
hydroxyphenylacetic acid)-N0-(p-hydroxyphenylacetic acid);
p,p-EDDHA, ethylenediamine-N-(p-hydroxyphenylacetic acid)-
N0-(p-hydroxyphenylacetic acid); EDTA, ethylenediaminetetraa-
cetic acid; MES, 2-(N-morpholino)ethanesulfonic acid; NMR,
nuclear magnetic resonance; AAS, atomic absorption spectros-
copy; IR, infrared spectroscopy; HRMS, high-resolution mass
spectrometry; ESI-MS, electrospray ionization mass spectrometry.
Constants; VCH Publishers: New York, 1992; 200 pp.
(17) Gans, P.; Savatini, A.; Vacca, A. Investigation of equilibria in
solution. Determination of equilibrium constants with the Hyper-
quad suite of programs. Talanta 1996, 43, 1739–1753.
(18) L’Eplattenier, F.; Murase, I.; Martell, A. New multidentate ligands,
VI: chelating tendencies of N, N0-di(2-hydroxybenzyl)ethylenedi-
amine-N,N0-diacetic acid. J. Am. Chem. Soc. 1967, 89, 837–843.
(19) Gustafsson, J. P. Visual MINTEQ Version 6.0 Kungl Tekniska
Hogskolan (KTH), Division of Land Water Resources, 2006;
updated June 21, 2006, http://www.lwr.kth.se/English/OurSoftware/
vminteq/.
Note Added after ASAP Publication
There was an error in the Introduction of the version of this
paper published ASAP on June 9, 2010; the correct version pub-
lished on June 11, 2010.
(20) Halvorson, A. D.; Lindsay, W. L. Equilibrium relationships of metal
chelates in hydroponic solutions. Soil Sci. Soc. Am. Proc. 1972, 36,
755–761.
LITERATURE CITED
(21) Pecoraro, V. L.; Bonadies, J. A.; Marrese, C. A.; Carrano, C. J.
Stepwise metal-assisted oxidative decarboxylation of vanadium(V)
ethylenebis((o-hydroxyphenyl)glycine). Isolation of a possible inter-
mediate. J. Am. Chem. Soc. 1984, 106, 3360–3362.
(22) Bonadies, J. A.; Carrano, C. J. Vanadium phenolates as models for
vanadium in biological systems. 1. Synthesis, spectroscopy, and elec-
trochemistry of vanadium complexes of ethylenebis[(o-hydroxy-
phenyl)glycine] and its derivatives. J. Am. Chem. Soc. 1986, 108,
4088–4095.
(23) Riley, P. E.; Pecoraro, V. L.; Carrano, C. J.; Bonadies, J. A.;
Raymond, K. N. X-ray crystallographic characterization of a step-
wise, metal-assisted oxidative decarboxylation: vanadium complexes
of ethylenebis[(o-hydroxyphenyl)glycine] and derivatives. Inorg.
Chem. 1986, 25, 154–160.
(24) Carrano, C. J.; Spartalian, K.; Appa Rao, G. V. N.; Pecoraro, V. L.;
Sundaralingam, M. The iron(III) complex of N-[2-((o-hydroxyphe-
nyl)glycino)ethyl]salicylidenimine. A model complex for the iron(III)
environment in the transferrins. J. Am. Chem. Soc. 1985, 107, 1651–
1658.
(1) Chaney, R. L.; Bell, P. F. Complexity of iron nutrition: lessons for
plant-soil interaction research. J. Plant Nutr. 1987, 10, 963–994.
(2) Chen, Y.; Barak, P. Iron nutrition of plants in calcareous soils. Adv.
Agron. 1982, 35, 217–240.
(3) Mengel, K.; Kirby, E. A.; Kosegarten, H.; Appel, T. Iron. In
Principles of Plant Nutrition; Mengel, K., Kirby, E. A., Eds.; Kluwer
Academic Publisher: Dordrecht, The Netherlands, 2001; pp 553-571.
(4) Nowack, B. Enviromental chemistry of aminocarboxylate chelating
agents. Environ. Sci. Technol. 2002, 36, 4009–4016.
(5) Nowack, B. Chelating agents and the environment. Environ. Pollut.
2008, 153, 1–2.
ꢀ
ꢁ
(6) Cremonini, M. A.; Alvarez-Fernandez, A.; Lucena, J. J.; Rombola,
´
A.; Marangoni, B.; Placucci, G. J. Nuclear magnetic resonance
analysis of iron ligand EDDHA employed in fertilizers. J. Agric.
Food Chem. 2001, 49, 3527–3532.
(7) Herna
´
ndez-Apaolaza, L.; Garcı
´
a-Marco, S.; Nadal, P.; Lucena, J. J.;
rez-Lopez, P.; Escudero, R.
Sierra, M. A.; Go
´
mez-Gallego, M.; Ramı
´
´
Structure and fertilizer properties of byproducts formed in the
(25) Li, X.; Pecoraro, V. L. Stepwise, metal-assisted decarboxylation
promoted by manganese: reactivity relationship between manganese
and vanadium. Inorg. Chem. 1989, 28, 3403–3410.
synthesis of EDDHA. J. Agric. Food Chem. 2006, 54, 4355–4363.
´ ´
(8) Laghi, L.; Alcaniz, S.; Cerdan, M.; Gomez-Gallego, M.; Sierra,
M. A.; Placucci, G.; Cremonini, M. A. Facile deferration of
commercial fertilizers containing iron chelates for their NMR
(26) Yunta, F.; Garcıa-Marco, S.; Lucena, J. J. Theoretical speciation of
´
ethylenediamine-N-(o-hydroxyphenylacetic)-N0-(p-hydroxyphenyl-
acetic) acid (o,p-EDDHA) in agronomic conditions. J. Agric. Food
Chem. 2003, 51, 5391–5399.
analysis. J. Agric. Food Chem. 2009, 57, 5143–5147.
ꢀ
(9) Orera, I.; Orduna, J.; Abadı
´
a, J.; Alvarez-Ferna
´
ndez, A. Electro-
spray ionization collision-induced dissociation mass spectrometry:
a tool to characterize synthetic polyaminocarboxylate ferric chel-
ates used as fertilizers. Rapid Commun. Mass Spectrom. 2010, 24,
109–119.
(27) Smith, R. M.; Martell, A. E. Critical Stability Constants; Plenum
Press: New York, 1984.
(28) Bannochie, C. J.; Martell, A. E. Affinities of racemic and meso forms
of N,N0-ethylenebis[2-(o-hydroxypheny1)glycine for divalent and
trivalent metal ions. J. Am. Chem. Soc. 1989, 111, 4735–4742.
(29) Ahrland, S.; Dahlgren, A.; Persson, I. Stabilities and hydrolysis of
some iron(III) and manganese(III) complexes with chelating ligands.
Acta Agric. Scand. 1990, 40, 101.
(10) Garcıa-Marco, S.; Martınez, N.; Yunta, F.; Hernandez-Apaolaza,
´ ´ ´
L.; Lucena, J. J. Effectiveness of ethylenediamine-N-(o-hydro-
xyphenylacetic)-N0-(p-hydroxyphenylacetic) acid (o,p-EDDHA) to
supply iron to plants. Plant Soil 2006, 279, 31–40.
(11) Yunta, F.; Garcıa-Marco, S.; Lucena, J. J. Chelating agents related
´
(30) Lindsay, W. L. Chemical Equilibrium in Soils; Wiley: New York, 1979.
´ ´
(31) Arias, M.; Lopez, E.; Fernandez, D.; Soto, B. Copper distribution
to ethylenediamine bis(2-hydroxyphenyl)acetic acid (EDDHA):
synthesis, characterization, and equilibrium studies of the free-
ligands and their Mg2þ, Ca2þ, Cu2þ, and Fe3þ chelates. Inorg. Chem.
2003, 42, 5412–5421.
and dynamics in acid vineyard soils treated with copper-based
fungicides. Soil Sci. 2004, 169, 796–805.
(12) Go
´
mez-Gallego, M.; Sierra, M. A.; Alca
zar, R.; Ramırez, P.; Pinar,
´ ´
Received for review March 15, 2010. Revised manuscript received
May 27, 2010. Accepted May 28, 2010. Financial support by Spanish
MCINN Grants AGL2007-63756 (UAM group), CTQ2007-67730-
C02-01, CAM CCG07-UCM/PPQ-2596, and Consolider-Ingenio
2010 CSD2007-0006 (UCM group) and Tradecorp is acknowledged.
R.E. thanks Tradecorp for financial support.
C.; Mancheno, J.; Garcı
´
a-Marco, S.; Yunta, F.; Lucena, J. J.
Synthesis of o,p-EDDHA and its detection as the main impurity in
o,o-EDDHA commercial iron chelates. J. Agric. Food Chem. 2002,
50, 6395–6399.
(13) Rojas, C. L.; Romera, F. J.; Alca
ntara, E.; Perez-Vicente, R.;
´ ´
Sariego, C.; Garcıa-Alonso, J. I.; Boned, J.; Marti, G. Efficacy of
´