New chelating ligands with potential for pharmaceutical applications

Article abstract of DOI:10.1080/15533170903492861

2-hydroxybenzylaminomethylphosphonic acid (HBAMPA) has been synthesized from commercially available chemicals and its aqueous coordination chemistry involving Al³⁺, Ga³⁺ and Fe³⁺ is described. At 25C and 0.1 M ionic strength KNO3, both Al³⁺and Ga³⁺ formed dinuclear complexes while Fe³⁺ formed the simple 1:1 complex (FeL) along with a mono-protonated complex (FeLH). HBAMPA formed neutral complexes with these trivalent metal ions, which lead to an early precipitation. The solution mixtures precipitated at pH 6.0 for Al³⁺, pH 4.0 for Fe³⁺, and pH 3.0 for the Ga³⁺ ion. The carboxylic acid analogue of HBAMPA, 2-hydroxybenzylalanine (HBA) also has been tested with Al³⁺ and Fe³⁺. It appeared that HBA solubilized the Al³⁺ ion at pH 7.4. However, the Fe³⁺-HBA solution mixture formed a complex of a very minimum solubility and an early pH precipitation prevailed. HBA and HBAMPA are multi-dentate, phenol-containing ligands capable of binding a variety of metal ions with very high stability constants. These stability constants are discussed. Copyright Taylor & Francis Group, LLC.

Full text of DOI:10.1080/15533170903492861

Synthesis and Reactivity in Inorganic, Metal-Organic, and Nano-Metal Chemistry, 40:45–50, 2010
Copyright © Taylor & Francis Group, LLC
ISSN: 1553-3174 print / 1553-3182 online
DOI: 10.1080/15533170903492861
New Chelating Ligands with Potential for Pharmaceutical
Applications
Yahia Z. Hamada and Wesley R. Harris
Department of Chemistry and Biochemistry, University of Missouri-St. Louis, St. Louis, MO, USA
ganisms upon the release of Al3+ ion from the soil by the effect
of acid rains.[2] Aluminum was long considered as an innocent
metal ion; only during the last three decades this metal ion has
been found responsible for bone and neurological disorders.[2−7]
The biological essentiality of aluminum is extremely limited in
the living organisms. Aluminum may activate succinic dehy-
drogenase and δ-aminolevulinate dehydrase that is involved in
the porphyrin and hemoglobin synthesis.[8] On the other hand,
living organisms will not survive without iron.[1] Although iron
is essential to life, excessive iron intake can cause siderosis
and damage to the organs through excessive iron storage or
hemochromatosis [1,8,9] so that designing new chelating agents
for both aluminum and iron overloads is of great importance for
chelation therapy. It is well established in the literature that the
phenolate moiety has high affinity for the Fe3+ ion forming ferric
complexes of high stablilities.[10−13] Herein, we are introducing
new ligands that can chelate trivalent metal ions in aqueous so-
lutions. The new ligands stability constants with iron, aluminum
and gallium were measured. The spectroscopic titrations of the
formed complexes were collected. Protonation constants for the
free ligands as well as stability constants expressed in the form
of log βs are presented.
2-hydroxybenzylaminomethylphosphonic acid (HBAMPA) has
been synthesized from commercially available chemicals and its
aqueous coordination chemistry involving Al3+, Ga3+ and Fe3+ is
described. At 25◦C and 0.1 M ionic strength KNO3, both Al3+ and
Ga3+ formed dinuclear complexes while Fe3+ formed the simple
1:1 complex (FeL) along with a mono-protonated complex (FeLH).
HBAMPA formed neutral complexes with these trivalent metal
ions, which lead to an early precipitation. The solution mixtures
precipitated at ∼ pH 6.0 for Al3+, ∼ pH 4.0 for Fe3+, and ∼ pH
3.0 for the Ga3+ ion. The carboxylic acid analogue of HBAMPA,
2-hydroxybenzylalanine (HBA) also has been tested with Al3+ and
Fe3+. It appeared that HBA solubilized the Al3+ ion at pH 7.4.
However, the Fe3+-HBA solution mixture formed a complex of a
very minimum solubility and an early pH precipitation prevailed.
HBA and HBAMPA are multi-dentate, phenol-containing ligands
capable of binding a variety of metal ions with very high stability
constants. These stability constants are discussed.
Keywords aluminum, chelating agents, iron, phosphonic acids
INTRODUCTION
Aluminum and iron are the most abundant metals in the
earth’s crust. The earth’s crust contains about 8.2% aluminum
by weight while the essential iron constitutes about 5.6% of the
weight of the earth’s crust.[1] Aluminum is more abundant than
most trivalent metal ions, yet nature did not utilize the Al3+ ion
as essential metal ion. One of the factors that may explain this
phenomenon is the fact that the aluminum ion possesses single
oxidation state of +3 while most of the other trivalent metals
possess redox chemistry. Aluminum is accessible to living or-
EXPERIMENTAL SECTION
Synthesis of HBAMPA
Aminomethylphosphonic acid was purchased from Aldrich
Chemical Company, Inc. (Milwaukee, WI). Salicyaldehyde and
sodiumborohydride were purchased from Fisher Chemical Co.
(Fair Lawn, NJ). Methanol was distilled prior to any reaction
over magnesium metal. Deuterium oxide was purchased from
Cambridge Isotope Laboratory.
Salicyaldehyde (5.47 mmol, 668 mg) in 25 mL methanol was
added dropwise within 10 minutes to a suspended solution of
aminomethylphosphonic acid (5.47 mmol, 607 mg) in 100 mL
Received 12 October 2009; accepted 10 November 2009.
The authors would like to thank the Research Board of the Univer-
sity of Missouri for financial support of this work (under Grant No.
S-3-40454). The authors would like to thank Dr. Janet Wilking of the
department of chemistry and biochemistry at UMSL for helping in the methanol in the presence of 2 equivalent of KOH (10.87 mmol,
data acquisition of NMR spectra and Dr. Mark M. Jones of the depart-
610 mg). The white suspension of aminomethylphosphonic acid
ment of chemistry and center in molecular toxicology at Vanderbilt
turned to a canary yellow solution because of the formation of
University for supplying samples of HBA.
the Schiff’s base. The mixture was refluxed overnight, then
Address correspondence to Yahia Z. Hamada, Division of Natural
allowed to cool before the addition of sodium borohydride as
reducing agent to obtain the free amine. It appeared that one
and Mathematical Science, LeMoyne-Owen College, 807 Walker Ave.,
Memphis, TN 38126. E-mail: wharris@umsl.edu
45

46
Y. Z. HAMADA AND W. R. HARRIS
equivalent of bororhydride was not sufficient so that a total of the free metal ions have been accounted for from Mesmer and
two equivalents of borohydride (11.1 mmol, 423 mg) was added Baes (See Table 5 for details).[16]
as solid. The mixture was allowed to stir overnight forming
a white suspension. The solid material was filtered through a
Buchner funnel. The solid was dissolved in minimum amount
of water; the initial pH of the solution was ≈ 10. The pH was
adjusted to ≈ 2.0 using 2 M HCl. Upon cooling the solution to
−4.0◦C, white micro crystals precipitated out of the solution.
The crystals were collected by filtration and dried under reduced
pressure for 72 hours.
Stability Constants Measurements
In brief, the typical titration system contains these basic com-
ponents, metal, ligand and hydrogen ion. Equation (1) shows the
general formula of a complexation reaction. The overall stabil-
ity constants for the formed complexes (βs) were expressed as
shown in equation (2).
The white crystals were characterized as follow: mp = 160
1◦C (dec.), 1H NMR referenced to D2O showed a peak at 3.0
ppm (d) of CH2 attached to phosphorus, a peak at 4.4 ppm (s)
CH2 of the benzylic hydrogens, two peaks at 7.0 ppm (m), and
7.4 ppm (m) two protons each of the aromatic ring. 13C NMR
showed the methylenic, the benzylic and the aromatic carbons
only at 46 ppm, 51 ppm, and (119–133) ppm, respectively. The
31P NMR peak appeared at 9.2 ppm referenced to phosphoric
acid.
iM + jL + kH → [MiLjHk]
[MiLj Hk]
[1]
[2]
βijk
=
[M]i[L]j [H]k
where MiLj Hk represent the different combination of the
formed complexes, M represents the free unhydrolyzed hex-
aaquo metal ion, L represents the uncomplexed, totally deproto-
nated form of the ligand, and H represents the free hydrogen ion
concentration, which is given by the pH value. Each complex
has one particular (β) value. The pH data points were fit with
different models where a model is a unique combination of dif-
ferent complexes that can form simultaneously using the least
squares data regression. The value of goodness of fit (GOF) as
given in equation (3) is the criterion that governs the success
of the proposed model. Under the potentiometric experimental
conditions, a value of 0.01 or less is considered as an acceptable
value.
Potentiometric Titration
Potentiometric titrations for free ligands, 1:1 and 1:2
metal:ligand systems were performed using a thermostated all
glass titration cell that was maintained at 25.0 0.1◦C by an
external circulating water bath. The 0.1 M KOH titrant solutions
were prepared from DILUT-IT ampoules purchased from J.T.
Baker Chemical Co and kept under an ascarite CO2 - scrubber in
the reservoir of a Metroham model 655 autoburette. The absence
of carbonate was confirmed for each KOH solution by Gran’s
plots.[14] Stock aluminum solutions were prepared by dissolv-
ing reagent grade aluminum chloride in a 100 mM hydrochloric
acid solution to prevent metal hydrolysis.
All titrations were carried out under argon atmosphere, the
ionic strength of all solutions was adjusted to 0.10 M by the ad-
dition of 1.0 M KNO3 solution. The pH was measured directly
with an accumet model 25 pH meter equipped with an accu
pHast combination electrode, both were purchased from Fisher
Scientific. The titrations were conducted using a computer-
controlled autotitrator. The autotitrator added the titrant aliquots,
stirred the component of the titration cell and monitored the pH
versus time. When the pH drift fell below 0.001 pH unit per
minute, the final pH was recorded and the burette was prompted
for the following addition of the titrant. In a typical experimental
run, the data points were collected throughout the pH range of
2.50 to 11.50. The equilibration time was preset and was long
enough for complete equilibration. A default preset timing was
chosen to be 1000 seconds. Any data point that exceeded the
preset time was discarded from the calculations. The concentra-
tions of the reactants were in the order of 0.001 M or 0.002M.
The titrations were conducted in which the metal:ligand ratios
were 1:1 and 1:2. Calculating the stability constants was done
using the least squares data fit that was described in details in
a previous report.[15] In all stability constants calculations for
Al3+, Fe3+, and Ga3+, four different hydrolysis constants for
ꢀ
pHobs − pHc2alc
GOF =
[3]
Nobs − Nv
where pHobs and pHcalc are the observed and calculated pH’s
respectively, Nobs is the number of observed data points, and Nv
is the number of variables in the proposed model.
The UV-VS spectra were collected using the computer con-
trolled cary-14 spectrophotometer at 25.0 0.1◦C, within the
concentration rang of 100 µM total ligand in 0.1 M KNO3 or
0.1 M KCl solutions. The IR spectra were collected by using
perkin-elmer 1600 FTIR spectrophotometer by preparing KBr
pellets. The KBr pellets were prepared from dry IR grade solid
from Aldrich Chemical Co.
RESULTS
Scheme 1 shows the new ligands introduced in this study.
Scheme 2 shows the synthetic routs taken to prepare HBAMPA.
SCH. 1. Chelating phenol-containing ligands used in this study.

NEW CHELATING LIGANDS
47
TABLE 1
Protonation constants and log βs for the different species
formed between Al3+ and HBAMPA, 25◦C, I = 0.10 M KNO3
# of protons
release
Species
Log β
pK a
σ
per aluminum
011 (HL)
012 (H2L)
013 (H3L)
223 (Al2L2H3)
112 (AlLH2)
111 (AlLH2(OH))
11.65
20.61
25.95
51.00
25.49
21.55
11.65 0.320
1.0
2.0
3.0
1.5
1.0
2.0
3.0
8.96
5.34
—
0.055
0.035
0.16
0.20
0.27
0.21
3.76
4.77
—
SCH. 2. Synthetic routes used to prepare the new HBAMPA from commer-
cially available compounds using water and methanol as solvents.
110 (AlH2L(OH)2) 16.73
The elemental analysis and the potentiometric measurements of
free HBAMPA indicated that there was one water molecule
of hydration associated with the free ligand. Figure 1 of
the Supplementary Material shows the potentiometric titration
Titrations of HBAMPA with Al3+
The potentiometric titrations of HBAMPA with Al3+ showed
unusual initial inflection point that appeared at 1.5 equiva-
lents of titrant per Al3+ and another major inflection at 3.0
equivalent. This initial break is observed and consistent by
titrating both 1:1 and 1:2 Al3+:HBAMPA. Such a half inte-
ger has been observed for the titrations of different relatives
of HBAMPA, aminoalkylphosphonic acids, aminomethylphos-
phonic acid and aminoethylphosphonic acid with aluminum.[15]
The position of such half integer inflection indicates that a bin-
uclear species is present. The Al3+-HBAMPA complex mixture
showed a precipitation around pH 6 0.5 with both 1:1 and 1:2
titrations.
curves of free HBAMPA and the titration curves of Al3+, Ga3+
,
and Fe3+ metal ions with an equivalent of HBAMPA. Free
HBAMPA has been defined as H3L ligand where the phospho-
nic acid group has a pKa of 5.34, the amine group has a pKa of
8.96, and the phenolic group has a high pKa of 11.65 (see Table
1). With the Al3+ ion the titration systems started to precipitate
around pH 6.0
0.5. With the Fe3+ ion the titration system
started to precipitate at pH 4.0 0.5, while with the Ga3+ ion
the titration systems started to precipitate at pH 3.0 0.5.
For Al3+-HBAMPA titration systems, it appeared that the
major complex formed is a one to one dinuclear complex in
which three protons remain undissociated. This complex was
designated as 223. The first index in the complex designation
stands for the number of the Al3+ ions, the second index stands
for the number of the HBAMPA ligand, and the third index
stands for the number of the hydrogen ions in the complex. The
223 was the complex that accounted for the position of the in-
flection at 1.5 equivalents of titrant per Al3+. The titration data
points were fit with a series of the simple one to one complexes.
The model that fit the potentiometric data points contained log
βs of 223, 112, 111, and 110 complexes. Alternative designa-
tions of these four complexes are (Al2L2H3, AlLH2, AlLH, and
AlL, respectively). Figure 1 is the speciation diagram of the
distribution of total aluminum among these complexes. It ap-
peared that the mono-protonated complex or 111 has not been
refined simultaneously with 223, 112, and 110 complexes when
the Al3+ to HBAMPA ratio is 1:1. The only way that the mono-
protonated complex (111) was observed and refined in the least
squares refinement within the 1:1 titration system is when the
log β value was kept as a constant parameter averaged from the
1:2 titrations and the other three complexes refined simultane-
ously. Such practice is not that uncommon in stability constants
measurements.[13−15]
FIG. 1. Speciation curves of the different complexes in the Al3+:HBAMPA
system with the 1:2 ratio as a function of pH: CAl = 0.01 M, CHBAMP A = 0.02
M. The complexes shown are the best model that fit the titration data points.

48
Y. Z. HAMADA AND W. R. HARRIS
The model that fit the 1:2 titration system contained 223,
112, 111, and 110 complexes all as variable parameters. All
complexes formed and refined simultaneously with the average
value of GOF of 0.0055. Figure 1 is the distribution diagram of
the total Al3+ among the different species that have been ob-
served within the 1:2 Al3+:HBAMPA titrations. Table 1 shows
the protonation constants of HBAMPA as well as log β-values
of the different Al-HBAMPA complexes. In Figure 1, the major
complex (110, AlL) has been maximized and fully formed about
pH 6, the other major complex or the dinuclear (223, Al2L2H3)
complex has been maximized and fully formed about pH 4 that
is well bellow the first hydrolysis constant of the Al3+ ion. The
diprotonated complex, or 112, has been formed at the very be-
ginning of the titration to about 50 % that was converted to the
monoprotonated (111, AlLH) up on the increase of the pH. The
monoprotonated 111 was built up to more than 20% about pH
5.8. Table 1 shows that the monoprotonated 111 complex has a
pKa of 4.82 that is about 0.64 log unit below the first hydrolysis
constant of the Al3+ ion.[16] It is worth noting here that the sim-
ple one to one complex or 110 has built up to about 90% below
pH 6.0 where the precipitation process has started. The simple
one to one complex 110 is neutral species with a net charge of
zero.
FIG. 2. Ultraviolet absorption spectrum of free HBA as function of pH-values,
TL = 1.02 × 10−4 M, at 25◦C.
Other models have been tested to fit the potentiometric data
points. Such models contained a series of dinuclear complexes
or 223, 222, 221, and 220. Any model based on such complexes
never showed least squares conversion but rather the programs
diverged. It looks like that the tri-protonated complex or 223
dissociates to both of 112, and 111 complexes.
equivalent of titrant per equivalent of Fe3+. Such position of
the inflection indicated that the simple one to one complex has
been formed. The half integer inflection that was observed with
Al3+ was missing. The three protons that have been titrated
could be those of the three functional groups of the ligand, the
phosphonic acid, the amine, and the phenolate functionalities.
There was a red precipitate that appeared around pH 4 0.5.
Titrations of HBAMPA with Ga3+
The Ga-HBAMPA system showed a major inflection at 3.0 The participation of the phenolic functionality in the chelation
however, the solution mixture did precipitate as early as pH 3.3 of Fe3+ within the acidic region can not be confirmed from
which allows a very narrow band of equivalents of titrant to work the potentiometric measurements so that we conducted UV-VS
with to define the complexes formed. By collecting the precip- measurements to see the effect of the chelation of the Fe3+
itate, drying it under vacuum, and looking at the IR spectrum, on the phenolic functionality within the acidic region. This is
the spectrum showed that the precipitate is not the amorphous shown in the supplementary Figure 2. Stability constants for
gallium hydroxide Ga(OH)3. The only observed species that Fe3+/ HBAMPA system are shown in the form of log βs in
has been formed with Ga3+ was the dinuclear tir-protonated Table 2.
complex or 223 complex just like the aluminum complex. The
gallium-HBAMPA complex had the stability value of (log β =
The Ligand 2-Hydroxybenzylalanine (HBA) and its
53.48 0.44) averaged from both the 1:1 and 1:2 titration sys-
Titrations with Fe3+
tems. Other model alternatives have been tested such as 223
Samples of HBA were received as a generous gift from Pro-
along with 112 or 111, or 110 or all of these complexes com-
fessor Mark M. Jones of the Department of Chemistry and
bined in one model. Also, other different combinations, such as
112, 111, and 110, were not successful models. All of the pro-
posed models have never converged, but they rather diverged
rejecting the formation of the simple one to one complexes such
as 110.
TABLE 2
Log βs for the different species formed between Fe3+ and
HBAMPA, 25◦C, I = 0.10 M KNO3
Species
Log β
σ
pKa
# of proton per iron
Titrations of HBAMPA with Fe3+
111 (FeLH)
110 (FeL)
24.39
20.90
0.17
0.29
3.49
2.0
3.0
The iron HBAMPA behaved differently compared to both
Al3+ and Ga3+, where the observed inflection appeared at 3.0

NEW CHELATING LIGANDS
49
TABLE 3
TABLE 4
Protonation constants and Log βs for the different species
Log βs for the different species formed between Al3+/ HBA,
25◦C, I = 0.10 M KNO3
formed between Fe3+/ HBA, 25◦C, I = 0.10 M KNO3
%
Formed
Number # of
of proton
Complex Log β
σ
% Formed Avg. fit Number of runs
11-1 8.53 0.31
88% 0.046 13
Complex Log β
σ
(pK a) Avg. fit runs per iron
011 (HL) 10.96 0.06 (10.96) 0.0047
012 (H2L) 19.62 0.03 (8.66)
013 (H3L) 21.96 0.03 (2.34)
3
1.0
2.0
3.0
2.0
3.0
4.0
or 11-1 complexes. Table 3 shows the protonation constants
along with the log β-values for the different species formed
between Fe3+/ HBA. Supplementary Figure 4 shows the visible
absorption spectra of Fe3+-HBA as a function of pH in which
the total ligand concentration (TL) = the total iron concentration
(TFe) = 1.05 × 10−4 M at 25◦C.
111
110
11-1
19.99 0.24 20% 0.0040
17.44 0.13 75%
14.18 0.11 20%
3
7
3
Center in Molecular Toxicology at Vanderbilt University in
Nashville, TN USA. HBA ligand has been defined as H3L lig-
and in which the carboxyl group has a pKa of 2.34, the amine
group has a pKa of 8.66 and the phenolate group has a pKa of
10.96. Log βs for the free ligand and for the HBA/Fe3+ system
is shown in Table 3.
Titration of HBA with Al3+
In contrast to the Fe3+/HBA titrations the Al3+/HBA titration
systems showed no visible precipitate at any pH value (Table 4).
Both 1:1 and 1:2 titratins showed the same pattern with initial
inflection at 1.0 equivalent of titrant per aluminum and a major
inflection at 4.0 equivalents of titrant per aluminum. Titrating
a net of four equivalents of protons with a major inflection at
4.0 equivalents may suggest the formation of the mono-hydroxo
complex or 11-1 complex. The only successful model that con-
verged the potentiometric data points was a model that contained
this mono-hydroxo complex or 11-1 complex only. Other alter-
native models were used that contained the mono-protonated
111 and the simple one to one 110 complexes. None of these
models were successful where the least squares fit diverged and
rejected the proposed log β values for both 111 and 110 com-
plexes indicating their absence.
With HBA:Fe3+ titrations, the system precipitated a brick
red like precipitate as early as pH 3.3. Supplementary Figure 3
shows the titration curves of the free HBA along with HBA:Fe3+
titration systems in 1:1 and 1:2 ratios. The major inflection
appeared at 4.0 equivalents of titrant per iron. This suggested that
the mono-hydroxo complex or 11-1 may be present. The model
that has been tested was that containing the mono-protonated
or 111, the simple one to one or 110, and the mono-hydroxo
Figure 2 is the ultraviolet absorption spectra for the free HBA
in its totally protonated form or H3L, the diprotonated form or
H2L in which the carboxylate group is de-protonated, and the
mono-protonated form or HL in which both the carboxylate
and the amine groups are de-protonated. The de-protonation of
the carboxylate showed no effect on the ultraviolet absorption
peaks. The only observed absorption peak is that for the phe-
nolate chromophore with the maximum absorption peak at 275
nm. The de-protonation of the amine group showed the distinct
peaks with maximum absorption at 237 nm and 292 nm with the
following molar extinction coefficients: ε237 = 6332 M−1Cm−1
and ε292 = 2781 M−1Cm−1
.
Figure 3 is the ultraviolet absorption spectra for the 1:1 reac-
tion mixture of Al3+ and HBA with a total Al3+ ion concentra-
tion of 102 µM. The reaction mixture had initial pH of 2.840.
NaOH solution with the concentration of 0.1 M was used to
increase the pH of the reaction mixture. The corresponding ab-
sorption peaks of the Al3+-HBA reaction mixture between pH
2.840 to pH 7.222 is more or less identical showing the absorp-
tion of the phenolate chromophore only. This suggested that the
amine group has no participation in the aluminum chelation.
FIG. 3. Ultraviolet absorption spectrum of Al3+-HBA reaction mixture as a
function of pH. TL = 1.02 × 10−4 M, at 25◦C.

50
Y. Z. HAMADA AND W. R. HARRIS
aqueous solutions.[13,16,18] For every metal used in this study we
have taken into account four different hydrolysis constants, as
shown in Table 5.
TABLE 5
The hydrolysis constants of the trivalent metal ions used in this
study in the form of log βs from Ref. 16
Metal/Species
Al3+
Fe3+
Ga3+
REFERENCES
10-1 (M(OH))
10-2 (M(OH)2)
10-3 (M(OH)3)
10-4 (M(OH)4)
20-2 (M2(OH)2)
−5.46
−10.04
−15.74
−23.49
—
−2.68
−6.41
—
−3.09
−6.64
−11.04
−17.09
—
1. Huheey, J.E., Keiter, E.A., and Keiter, R.L. Inorganic Chemistry, Princi-
ples of Structure and Reactivity, 4th ed., New York, NY: Harper Collins
Publishers, 1993.
−22.09
2. Martin, R.B. Aluminum: a neurotoxic product of acid rain. Acc. Chem. Res.
1994, 27, 204–210.
3. Alfrey, A.C. Aluminum and renal disease. Contrib. Nephrol. 1993, 102,
110–124.
−2.95
4. McLachlan, D.R.C., Lukiw, W.J., and Kruck, T.P.A. New evidence for an
active role of aluminum in Alzheimer’s disease. Can. J. Neur. Sci. 1989,
16, 490–497.
5. McLachlan, D.R.C., Lukiw, W.J., Kruck, T.P., and Krishnan, S.S. Would de-
creased aluminum ingestion reduces the incidence of Alzheimer’s disease?
Can. Med. Assoc. J. 1991, 145, 793–804.
6. Ganrot, P.O. Metabolism and possible health effects of aluminum. Environ.
Health. Perspect. 1986, 65, 363–441.
7. Harris, W.R., Berthon, G., Day, J.P., Exley, C., Flaten, T.P., Forbes, W.F.,
Kiss, T., Orvig, C., and Zatta, P. Speciation of aluminum in biological
systems. Journal of Toxicology and Environmenal Health 1996, 48, 543–
568.
8. Bowen, H.J.M. Environmental Chemistry of The Elements. Academic: New
York, 1979.
9. Rollinson, C.L., and Ening, M.G. In: Kirk-Othmer Encyclopedia of Chem-
ical Technology, 3rd ed. Grayson, M. (Ed.), New York: Wiley, 1981, Vol.
15, pp 570–603.
DISCUSSION
The best description of the simple one to one complexes
formed between Al3+ and HBAMPA is where the phosphonate
moiety is coordinated to Al3+ while both the amine and the phe-
nolate are still protonated. The diprotonated complex or 112 is
best described as AlLH2 while the mono-protonated 111 com-
plex is best described as AlLH2(OH). So that the simple one to
one complex or 110 complex is best formulated as AlLH2(OH)2
and not simply AlL (this is the neutral metal complex with a net
zero charge that precipitated with the metal ions under consid-
eration). For the major dinuclear complex or 223, the bulkiness
of the two phenolate moieties in one molecular complex setting
is the factor that leads to the dissociation of the dinuclear com-
plex to two monomers of 112, and 111 complexes. The simple
stoichiometric addition of both 112 plus 111 will lead to 223.
The UV studies proved the absence of the participation of
both the amine and the phenolate functionalities of HBAMPA
from the Al3+ coordination. Equation (4) and (5) shows the
successive dissociation of the two protons from the diprotonated
112 complex.
10. Harris, W.R., Motekaitis, R.J., and Martell, A.E. New mutidentate lig-
ands XVII. Chelating tendencies of N-(o-Hydroxybenzyl)iminodiacetic
acid (H3L). Inorg. Chem. 1975, 14, 974–978.
11. Motekaitis, R.J., Sun, Y., and Martell, A.E. N,Nꢀ-Bispyridoxyle
thylenediamine-N,N’-diacetic acid (PLED) and N,Nꢀ-bis(2-hydroxy-
5-sulfobenzylethyelnediamine-N,N’-diacetic acid (SHBED). Inorganica
Chimica Acta 1989, 159, 29–39.
12. Ma, R., Motekaitis, R.J., and Martell, A.E. Synthesis of N-
hydroxybenzylethelene-N,NꢀNꢀ-triacetic acid and the stabilities of its com-
plexes with divalent and trivalent metal ions. Inorganica Chimica Acta
1995, 233, 137–143.
13. Martell, A.E., Smith, R.M., and Motekaitis, R.J. Critical Stability Constants
Database, Version 4.0, NIST, Texas A & M University, College Station,
TX, USA,1997.
14. Rossotti, F.J.C., and Rossotti, H. Potentiometric titrations using Gran plots
A textbook omission. J. Chem. Ed. 1965, 42, 375–378.
15. Hamada, Y.Z., and Harris, W.R. Stability constants of aluminum with phos-
phonic acid derivatives and multinuclear NMR-measurements in aqueous
solutions. Inorg. Chim. Acta. 2006 359(4) 1135–1146.
16. Baes, C.F., and Mesmer, R.E. The Hydrolysis of Cations. Wiley and Sons:
New York, 1967.
17. Harris, W.R., Wang, Z., and Hamada, Y.Z. Competition between transferrin
and the serum ligands citrate and phosphate for the binding of aluminum.
Inorg. Chem. 2003 42, 3262–3273.
[AlLH2]2+ → [AlLH2(OH)]+ + H+
[4]
[5]
[AlLH2(OH)]+ → [AlLH2(OH)2]0 ↓ ppt + H+
Equation (4) has an equilibrium constant Keq of 103.94 , while
equation (5) has as equilibrium constant Keq of 104.82. The log
β value for the tri-protonated 223 complex is such a high value
that reflects the complexity of the Al3+-HBAMPA equilibrium
process. Comparing this log β value of 51.00 with the analo-
gous log β value of ∼ 17.01 for the trimeric species that have
been observed in the Al3+-citrate system shows similar com-
plexity of aluminum equilibria in aqueous solutions.[17] The
chemistry of gallium is yet more complex compare to both Fe3+
and Al3+ due to the tendency of Ga3+ to hydrolyze to form
the gallate anion or Ga(OH)−4 .[16,18] Perhaps this is one of the
reasons for the enthusiasm of many reports of Ga3+ work in
18. Martell, A.E., and Szpoganicz, B. Zinc(II), aluminum(III), and gal-
lium(III) of Schiff bases of 3-Amino-3-phosphonopropionic Acid and 5ꢀ-
deoxypyridoxal. Inorg. Chem. 1989, 28, 4199–4206.

Copyright of Synthesis & Reactivity in Inorganic, Metal-Organic, & Nano-Metal Chemistry is the property of
Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv
without the copyright holder's express written permission. However, users may print, download, or email
articles for individual use.