Structure of concentrated aqueous urea solutions involving alkali metal salts studied by neutron diffraction with 14n/15n, 6li/7li, and 35Cl/37Cl isotopic substitution methods

Article abstract of DOI:10.1246/bcsj.20120225

Neutron diffraction with ₁₄N/₁₅N, 35Cl/37Cl, and 6Li/7Li isotopic substitution methods have been applied to aqueous 10 mol% urea solutions involving 5mol% NaCl and 10mol% LiCl, in order to obtain information on the hydration structure around the amino group of urea which is affected by coexisting ions, Cl and Li+. Observed first-order difference functions, N inter(Q), Cl(Q), and Li(Q), were respectively analyzed using least-squares fitting. The nearest neighbor intermolecular NOw (Ow: oxygen atom in D2O) distance was determined to be 3.09(1) and 3.19(1), from the analyses for sample solutions involving NaCl and LiCl, respectively. These intermolecular NOw distances are slightly longer than those reported for aqueous urea solutions of similar concentrations. These results suggest that hydrogen bonds among amino groups of urea and neighboring water molecules are weakened by coexisting alkali metal salts. The structural parameters concerning the first hydration shell of Cl- and Li+, have been determined to be r(ClDw) = 2.27(1) (Dw: deuterium atom in D₂O) and r(Li+Ow) = 1.95(1), respectively. The results indicate that the structure of the first hydration shell around Cl- and Li+ is not affected by the presence of urea molecules in the aqueous solution.

Full text of DOI:10.1246/bcsj.20120225

1
04
Bull. Chem. Soc. Jpn. Vol. 86, No. 1, 104­111 (2013)
© 2013 The Chemical Society of Japan
Structure of Concentrated Aqueous Urea Solutions Involving Alkali
1
4
15
6
7
Metal Salts Studied by Neutron Diffraction with N/ N, Li/ Li,
and Cl/ Cl Isotopic Substitution Methods
35
37
1
2
2
2
Takuya Miyazaki, Yasuo Kameda,* Yuko Amo, and Takeshi Usuki
1Graduate School of Science and Engineering, Yamagata University, Yamagata 990-8560
2Department of Material and Biological Chemistry, Faculty of Science, Yamagata University, Yamagata 990-8560
Received August 19, 2012; E-mail: kameda@sci.kj.yamagata-u.ac.jp
Neutron diffraction with 14N/ N, Cl/ Cl, and Li/ Li isotopic substitution methods have been applied to aqueous
15
35
37
6
7
1
0 mol % urea solutions involving 5 mol % NaCl and 10 mol % LiCl, in order to obtain information on the hydration
¹
+
structure around the amino group of urea which is affected by coexisting ions, Cl and Li . Observed first-order difference
functions, ¦N (Q), ¦Cl(Q), and ¦Li(Q), were respectively analyzed using least-squares fitting. The nearest neighbor
inter
intermolecular N£Ow (Ow: oxygen atom in D2O) distance was determined to be 3.09(1) and 3.19(1) ¡, from the analyses
for sample solutions involving NaCl and LiCl, respectively. These intermolecular N£Ow distances are slightly longer than
those reported for aqueous urea solutions of similar concentrations. These results suggest that hydrogen bonds among
amino groups of urea and neighboring water molecules are weakened by coexisting alkali metal salts. The structural
¹
+
¹
parameters concerning the first hydration shell of Cl and Li , have been determined to be r(Cl £Dw) = 2.27(1) ¡ (Dw:
+
deuterium atom in D2O) and r(Li £Ow) = 1.95(1) ¡, respectively. The results indicate that the structure of the first
hydration shell around Cl and Li is not affected by the presence of urea molecules in the aqueous solution.
¹
+
Urea is one of the simplest biomolecule and involves two
bonded H­H distance compared to pure water. In order to
determine hydration structure around the carbonyl group of the
functional groups, amino- (NH2) and carbonyl (C=O) groups.
Urea is known to be highly soluble in water and its concentrated
solution works as a strong protein denaturant. Because of these
characteristic properties and its important role in metabolism, it
is necessary to understand the hydration structure of molecular
urea. The hydration structures around the amino group of urea
have widely been investigated by neutron diffraction measure-
1
2
13
urea molecule, neutron diffraction measurements with C/ C
isotopic substitution were carried out, and revealed that the
9
hydration number around carbonyl group is, 4.3(3). In real
biological systems, urea plays an important role as a carrier of
waste nitrogen atoms through the urea cycle. Alkali metal and
chloride ions are commonly present in living systems. However,
the effect of these ions on the hydration structure of urea has
not yet been investigated by neutron diffraction. In order to
obtain information on local structure around the amino group of
1
4
15
1­5
ments with N/ N isotopic substitution. Finney et al. first
reported the distribution function around the nitrogen atom of
amino group, G (r), for 2.0 molal (3.8 mol %) urea solutions.
N
1
4
15
The number of D O molecules in the first hydration shell of urea
the urea molecule, neutron diffraction with N/ N isotopic
substitution is one of the most suitable experimental techniques.
2
was roughly estimated to be 7.1(5) through an integration of the
1,2
¹
+
G (r) from r = 2.55 to 4.00 ¡. The intermolecular interaction
Details of the local structure around coexisting Cl and Li ions
N
3
5
37
6
7
in the concentration range of 2.0 to 14.0 molal (3.8­22.2 mol %)
can also be determined by using Cl/ Cl and Li/ Li sub-
stituted sample solutions.
3
urea were examined by Turner et al. The observed G (r) over
N
a wide range suggests remarkably little change in the inter-
molecular structure around the amino group. Recently, studies
using a combination of neutron diffraction and empirical poten-
In the present paper, we describe results of neutron dif-
fraction measurements on concentrated 10 mol % urea heavy
water solutions involving 5 mol % Na*Cl and 10 mol % *LiCl
6
,7
14
15
35
37
6
7
tial structure refinement (EPSR) analysis have been reported.
in which isotopic ratios, N/ N, Cl/ Cl, and Li/ Li, have
The results of neutron diffraction measurements for different
isotopic ratios of H/D and 14N/ N of 20 mol % urea solu-
tions indicated that the hydration numbers around amino-
and carbonyl groups of the urea molecule are 3.16 and 2.49,
been changed. Observed intermolecular difference functions,
15
inter
¦
(Q), ¦ (Q), and ¦ (Q), were analyzed by least-squares
Cl Li
N
fitting to determine structural parameters concerning the local
environment around substituted atoms.
6
respectively. Results for supersaturated 30 mol % urea solution
7
Experimental
revealed that the hydration number of urea is ca. 6.1. The H­H
1
5
and H­(O, C, N) partial distribution function was obtained
from neutron diffraction measurements with H/D isotopically
Materials.
Isotopically enriched ( NH ) C=O (98.0%
2 2
1
5
N, Isotec Inc.) was deuterated by dissolving into 20 times the
8
substituted 10 molal (15.3 mol %) urea solutions. These partial
molar quantity of D O (99.9% D, Aldrich Chemical Co., Inc.),
2
distribution functions indicated a shift in the average hydrogen-
followed by dehydration under vacuum. This procedure was
Published on the web December 29, 2012; doi:10.1246/bcsj.20120225

T. Miyazaki et al.
Bull. Chem. Soc. Jpn. Vol. 86, No. 1 (2013)
105
Table 1. Isotopic Compositions, Total Cross Sections, and Number Density of the Stoichiometric Unit, (Na*Cl)0.05[(*ND2)2-
C=O]0.1(D2O)0.85 and (*LiCl)0.1[(*ND2)2C=O]0.1(D2O)0.8, ·t, and μ, Respectively
·
barns
t
μ
/¡
Sample
14N%
15N%
35Cl%
37Cl%
6Li%
7Li%
a)
¹3
/
I (Na Cl)0.05[(natND2)2C=O]0.1(D2O)0.85
nat
99.6
99.6
2.0
0.4
0.4
98.0
75.8
99.4
75.8
24.2
0.6
24.2
®
®
®
®
®
®
18.039
18.595
16.578
35
nat
II (Na Cl)0.05[( ND2)2C=O]0.1(D2O)0.85
0.02877
0.02826
nat
15
III (Na Cl)0.05[( ND2)2C=O]0.1(D2O)0.85
IV ( LiCl)0.1[(natND2)2C=O]0.1(D2O)0.8
6
99.6
99.6
2.0
0.4
0.4
98.0
®
®
®
®
®
®
95.4
0.1
0.1
4.6
99.9
99.9
73.576
19.750
18.289
7
nat
V ( LiCl)0.1[( ND2)2C=O]0.1(D2O)0.8
7
15
VI ( LiCl)0.1[( ND2)2C=O]0.1(D2O)0.8
a) For incident neutron wavelength of 1.09 ¡.
1
5
repeated 3 times to obtain ( ND2)2C=O with 99.8% D.
(
vanadium rod (10 mm in diameter), empty cell, and instrumen-
tal background.
nat
14
ND ) C=O (99.6% N and 98.0% D, Isotec Inc.) was
2 2
deuterated by dissolving into 10 times the molar quantity of
Data Reduction. Observed scattering intensities from the
sample solution were corrected for instrumental background,
D O (99.9% D, Aldrich Chemical Co., Inc.), followed by
2
1
0
11
removal of water under vacuum. This procedure was repeated 3
absorption, and multiple scattering. The observed count rate
for the sample solution was converted to the absolute scale by
the use of corrected scattering intensities from the vanadium
rod. Details of data correction and normalization procedures are
nat
times to obtain ( ND ) C=O with 99.8% D. In order to obtain
2
2
isotopically enriched 6LiCl, LiCO (95.3% Li, Tomiyama
6
6
3
Chemical Co., Ltd.) was dissolved in concentrated aqueous
hydrochloric acid solution (Nacalai tesque, guaranteed grade)
5
,12,13
given elsewhere.
The first-order difference functions,14­16 ¦ (Q) and ¦ (Q),
7
7
35
and dried under vacuum. LiCl (99.9% Li, Isotec Inc.), Na Cl
(
(
N
Cl
3
5
99.4% Cl, Aldrich Chemical Co., Inc.), and natural NaCl
Nacalai Tesque, guaranteed grade) were used without further
were obtained as the numerical difference in normalized
scattering cross section from urea solutions involving NaCl.
purification. Compositions of NaCl and LiCl were chosen to be
and 10 mol %, respectively, considering the solubility limit of
obs
ꢀ
ꢀ
NðQÞ ¼ ðd·=dꢀÞ ðfor sample IÞ
5
obs
ꢀ
ðd·=dꢀÞ ðfor sample IIIÞ
ð1Þ
ð2Þ
the urea­MCl­D O ternary system. Solute concentration of the
2
obs
ClðQÞ ¼ ðd·=dꢀÞ ðfor sample IIÞ
present sample solutions are much higher than that occurring
in real biological systems, however, the effect of the coexisting
ions on the hydration structure of the urea molecule is expected
to appear more precisely. The required amount of enriched
obs
ꢀ ðd·=dꢀÞ ðfor sample IÞ
Similarly, difference functions, ¦ (Q) and ¦ (Q), were
N
Li
obtained from the difference in scattering cross section from
compounds were dissolved into D O (99.9% D, Aldrich
sample solutions involving LiCl.
2
Chemical Co., Inc.) to prepare six kinds of aqueous 10 mol %
urea solutions involving alkali metal salts (NaCl or LiCl) with
different isotopic compositions. The density of sample solu-
tions (all natural abundance) was picnometrically determined to
be 1.156(2) and 1.159(1) g cm for the LiCl­urea and NaCl­
urea solutions, respectively. Each sample solution was sealed
in a cylindrical fused quartz cell (14.0 mm in an outer diameter
and 1 mm in wall thickness). The sample parameters used in the
present study are listed in Table 1.
Neutron Diffraction Measurements. Neutron diffraction
measurements were carried out at 25 °C using the ISSP dif-
fractometer 4G (GPTAS) installed at the JRR-3M reactor
operated at 20 MW in the Japan Atomic Energy Agency
obs
ꢀ
ꢀ
NðQÞ ¼ ðd·=dꢀÞ ðfor sample VÞ
obs
ꢀ
ðd·=dꢀÞ ðfor sample VIÞ
ð3Þ
ð4Þ
obs
LiðQÞ ¼ ðd·=dꢀÞ ðfor sample IVÞ
¹
3
obs
ꢀ ðd·=dꢀÞ ðfor sample VÞ
Difference functions, ¦ (Q), ¦ (Q), and ¦ (Q), scaled at
N
Cl
Li
the stoichiometric unit, (MCl) [(ND ) C=O] (D O) (M =
1¹x¹y
x
2 2
y
2
Na and Li), are expressed as the weighted sum of six partial
structure factors relating to the isotopically substituted X atom
(X = N, Cl, and Li),
ꢀ
XðQÞ ¼ A½aXOðQÞ ꢀ 1ꢁ þ B½aXDðQÞ ꢀ 1ꢁ
þ C½aXCðQÞ ꢀ 1ꢁ þ D½aXNðQÞ ꢀ 1ꢁ
þ E½aXMðQÞ ꢀ 1ꢁ þ F½aXClðQÞ ꢀ 1ꢁ
ð5Þ
(JAEA), Tokai, Japan. The incident neutron wavelength, ­ =
1
.090(3) ¡, was determined by Bragg reflections from KCl
where
powder. Beam collimations were 40¤­80¤­80¤ in going from the
reactor to the detector. The aperture of the collimated beam was
A ¼ 4ð1ꢀxÞyðb ꢀb15 ÞbO; B ¼ 8ð1ꢀ xꢀyÞðb ꢀb15NÞbD;
natN
natN
N
2 2 2 2
C ¼ 4y ðbnat ꢀb15 ÞbC; D ¼ 4y ðbnat ꢀb15 ÞbN;
N
N
N
N
2
0 mm in width and 40 mm in height. Scattered neutrons were
E ¼ 4xyðb ꢀb15 ÞbM; and F ¼ 4xyðbnat ꢀb15NÞbCl
natN
(5¤)
N
N
collected over the angular range 3.0 ¯ 2ª ¯ 118°, correspond-
ing to scattering vector magnitude range of 0.30 ¯ Q ¯
1
4
15
for solutions with different N/ N composition,
¹
1
9
.88 ¡ (Q = 4³ sin ª/­). Angular step intervals were chosen
A ¼ 2xð1ꢀxÞðb35 ꢀb ÞbO; B ¼ 4xð1ꢀxþyÞðb35 ꢀbnatClÞbD;
natCl
Cl
Cl
to be ¦2ª = 0.5° in the range 3.0 ¯ 2ª ¯ 40° and ¦2ª = 1.0°
in the range 41 ¯ 2ª ¯ 118°, respectively. The preset time was
C ¼ 2xyðb35 ꢀbnat ÞbC; D ¼ 4xyðb35 ꢀbnat ÞbN;
Cl
Cl
Cl
Cl
2
2
2
2
E ¼ 2x ðb35 ꢀb ÞbNa; and F ¼ x ðb35 ꢀbnat ÞbCl (5¤¤)
natCl
Cl
Cl
Cl
1
50 s for samples III and VI, and 200 s for sample I, II, IV, and
V, respectively. Scattering intensities were measured for the
for 35Cl/ Cl substituted urea­Na*Cl solutions, and
37

1
06
Bull. Chem. Soc. Jpn. Vol. 86, No. 1 (2013)
Structure of Concentrated Urea Solutions
Table 2. Values of the Coefficients of aij(Q) in eq 5
Difference functions
A/barns
B/barns
C/barns
D/barns
E/barns
F/barns
¦N(Q) (I­III)
¦Cl(Q) (II­I)
¦N(Q) (V­VI)
¦Li(Q) (IV­V)
0.0631
0.0112
0.0598
0.0416
0.1601
0.0283
0.1543
0.1061
0.0076
0.0013
0.0076
0.0053
0.0181
0.0038
0.0181
0.0149
0.0021
0.0004
¹0.0025
¹0.0001
0.0055
0.0011
0.0110
0.0076
ꢀNmodelðQÞ
A ¼ 2xð1 ꢀ xÞðb
6
ꢀ b
7
ÞbO; B ¼ 4xð1 ꢀ x þ yÞðb
6
ꢀ b
7
LiÞbD;
(5¤¤¤)
Li
Li
Li
C ¼ 2xyðb
6
ꢀ b
7
ÞbC; D ¼ 4xyðb
ÞbLi; and F ¼ 2x ðb
6
ꢀ b
7
ÞbN;
2
2
Li
Li
Li
Li
¼ ꢁ4yn ðb ꢀ b15 Þb expðꢀl Q =2Þ
N¡ natN
N
¡
N¡
2
2
2
2
E ¼ x ðb
6
ꢀ b
7
6
ꢀ b
7
LiÞbCl
Li
Li
Li
ꢂ sinðQr Þ=ðQrN¡Þ
þ 4³μðA þ B þ C þ D þ E þ FÞ expðꢀl0 Q =2Þ
N¡
6
7
2
2
for Li/ Li substituted urea­*LiCl solutions. The weighting
factors A ¹ F in eq 5 are numerically listed in Table 2. Since
magnitude of A and B are much larger than that for the
other coefficients, observed ¦ (Q) mainly involves structural
information on distribution of D O molecules around the
ꢀ3
ꢂ ½Qr0 cosðQr0Þ ꢀ sinðQr0ÞꢁQ
ð9Þ
model
ꢀ
Cl
ðQÞ
X
2
2
¼
ꢁ2xnCl¡ðb35 ꢀ b Þb¡ expðꢀlCl¡ Q =2Þ
natCl
Cl
2
ꢂ sinðQrCl¡Þ=ðQrCl¡Þ
X atom.
2
2
Observed difference function ¦ (Q) involves contributions
þ 4³μðA þ B þ C þ D þ E þ FÞ expðꢀl Q =2Þ
N
0
from both intra- and intermolecular interactions. These inter-
actions overlap each other in the interatomic distance range
of r < 3.5 ¡, where the nearest neighbor N£OW and N£DW
distances should fall. In the present analysis, intramolecular
interference contribution, IN (Q), was evaluated using known
molecular geometry of the urea molecule. IN (Q), consists of
ꢀ3
ꢂ ½Qr0 cosðQr0Þ ꢀ sinðQr0ÞꢁQ
ð10Þ
model
ꢀ
Li
ðQÞ
ꢁ2xnLi¡ðb
2
2
¼
6
ꢀ b Þb¡ expðꢀlLi¡ Q =2Þ
7
Li
Li
intra
ꢂ sinðQrLi¡Þ=ðQrLi¡Þ
intra
2
2
þ 4³μðA þ B þ C þ D þ E þ FÞ expðꢀl Q =2Þ
0
the sum of contributions from all N£¡ pairs within a urea
molecule.
ꢀ3
ꢂ ½Qr0 cosðQr0Þ ꢀ sinðQr0ÞꢁQ
ð11Þ
where, nX¡ (X = N, Cl, and Li), denotes the coordination
number of ¡ atom around the X atom. The long-range
parameter, r0, corresponds to the distance beyond which
a continuous distribution of atoms around the X atom can be
INintraðQÞ ¼ ꢁ4yðbnatN ꢀ b15NÞb¡ expðꢀlN¡ Q =2Þ
2
2
ꢂ sinðQrN¡Þ=ðQrN¡Þ
ð6Þ
where, lN¡ and rN¡ denote the root-mean-square displacement
and interatomic distance for the N£¡ pair, respectively. Values
of rN¡ and lN¡ used for the present analysis were taken from the
literature determined by the single crystal neutron diffraction
work and from values calculated for related molecules,
respectively. The calculated IN (Q) was then subtracted from
assumed. The parameter, l , describes the sharpness of the
0
boundary at r . Structural parameters, nX¡, lX¡, rX¡, r , and l are
0
0
0
respectively determined from the least-squares fit to the
1
7
18,19
observed ¦ (Q). The fitting procedure was performed in the
X
intra
¹1
23
range of 0.3 ¯ Q ¯ 9.88 ¡ with SALS program, assuming
that the statistical uncertainties distribute uniformly.
observed total ¦ (Q) to obtain the intermolecular difference
N
inter
function, ¦N (Q),
Results and Discussion
inter
intra
ꢀ
N
ðQÞ ¼ ꢀNðQÞ ꢀ IN ðQÞ
ð7Þ
Hydration Structure around the Nitrogen Atom of Urea
obs
The distribution function, G (r), describing the structure
Molecule. Observed scattering cross sections, (d·/dΩ) , for
X
around the isotopically substituted X atom is obtained by the
aqueous 10 mol % urea solutions involving 5 mol % NaCl with
Fourier transform of the observed ¦ (Q),
different isotopic compositions are shown in Figures 1a­1c.
X
ꢀ
1
obs
(
d·/dΩ)
for aqueous 10 mol % urea solutions involving
GXðrÞ ¼ 1 þ ðA þ B þ C þ D þ E þ FÞ
Z
Qmax
10 mol % LiCl are represented in Figures 1d­1f, respectively.
2
ꢀ1
14
15
35
37
ꢂ ð2³ μrÞ
QꢀXðQÞ sinðQrÞdQ
Although sample solutions have different N/ N, Cl/ Cl,
0
6
7
or Li/ Li isotopic compositions, the overall feature of
¼
½AgXOðrÞ þ BgXDðrÞ þ CgXCðrÞ þ DgXNðrÞ
þ EgXMðrÞ þ FgXClðrÞꢁ
obs
observed (d·/dΩ)
for each sample looks very similar,
difference functions, ¦ (Q), ¦ (Q), and ¦ (Q), were suc-
N
Cl
Li
ꢀ
1
ꢂ ðA þ B þ C þ D þ E þ FÞ
ð8Þ
cessfully derived from the difference between observed cross
sections by using eqs 1­4. The calculated intramolecular con-
¹
1
The upper limit of the integral, Q , was set to be 9.88 ¡ .
max
inter
intra
The intermolecular distribution function, GN (r), was eval-
tribution, IN (Q), was subtracted from the observed differ-
inter
uated by the Fourier transform of ¦N (Q).
ence function, ¦ (Q) (Sample I­III and V­VI), in order to
N
inter
Hydration parameters around N, Cl, and Li atoms were deter-
mined through least-squares fitting for observed ¦N (Q),
obtain the intermolecular difference function, ¦N (Q). The
present ¦N (Q) are characterized by resolved first diffraction
peak at 2 ¡ (Figures 2a and 2c).
inter
inter
model
¹1
¦
(Q), and ¦ (Q), employing the model functions ¦
(Q),
Cl
Li
N
model
model
¦
(Q), and ¦
(Q), involving both the short- and long-
Intermolecular distribution functions around the nitrogen
Cl
Li
range contributions:2
0­22
atom of the urea molecule, GN (r), are shown in Figures 3a
inter

T. Miyazaki et al.
Bull. Chem. Soc. Jpn. Vol. 86, No. 1 (2013)
107
Figure 2. a) Intermolecular diffraction function, ¦Ninter(Q),
observed for aqueous 10 mol % urea solutions involving
Figure 1. Scattering cross section, (d·/dΩ)obs, observed
for aqueous 10 mol % urea solutions involving NaCl or
5
mol % NaCl (dots), and the best-fit of the calculated
calc
nat
nat
¦N (Q) (solid line). c) Intermolecular difference func-
tion, ¦Ninter(Q), observed for aqueous 10 mol % urea
solutions involving 10 mol % LiCl (dots), and the best-fit
LiCl with different isotopic ratio of (a)
N­ Cl,
nat
35
15
nat
nat
6
nat
7
(
b) N­ Cl, (c) N­ Cl, (d) N­ Li, (e) N­ Li, and
15
7
(
f) N­ Li.
calc
of the calculated ¦N (Q) (solid line). b), d) Residual
functions (dots).
and 3b (indicated by dots). The first peak appearing at r µ
inter
3
.5 ¡ in the present GN (r) functions can be assigned to the
intermolecular interactions between the N atom and neighbor-
ing water molecules. The overall features of the present
treated as an independent parameter. The geometry of D O was
2
fixed to that reported for pure liquid heavy water (rOD =
inter
24,25
GN (r) functions for urea­NaCl­D O and urea­LiCl­D O
0.983 ¡, rDD = 1.55 ¡
). c) In the present analysis, the
2
2
solutions seem very similar to those previously reported for
second and third nearest neighbor N£D O interactions were
2
4
5
model
1
5 and 25 mol % aqueous urea solutions. This implies that a
taken into account for evaluating the theoretical ¦N
(Q)
stable first hydration shell is present around the amino group of
the urea molecule. The first intermolecular peak observed for
in order to improve the fit in the low-Q region. Each interaction
was treated as a single interaction with the value of b in eq 9
¡
the urea­LiCl­D O solution is significantly larger than that for
being 2b + b . d) Structural parameters for the long-range
2
D
O
the urea­NaCl­D O peak, suggesting that the first hydration
random distribution of atoms, l and r , were allowed to vary
0 0
2
+
shell of the amino group is more enhanced by coexisting Li
independently.
¹
The best-fit results of the observed ¦Ninter(Q) for urea solu-
tions involving NaCl and LiCl are shown in Figures 2a and 2c
(indicated by solid lines). Satisfactory agreement is obtained in
and/or Cl .
In order to obtain detailed information concerning the
hydration structure around the nitrogen atom of urea, the least-
squares fitting was applied to the intermolecular difference
¹
1
the range of 0.30 ¯ Q ¯ 9.88 ¡ (Figures 2b and 2d). Fourier
inter
model
function, ¦N (Q), observed for aqueous urea solutions
transform of the calculated ¦
(Q) was in good agreement
N
inter
involving NaCl and LiCl, respectively. In the present fitting
procedure, the following assumptions were adopted in evaluat-
ing theoretical interference function. a) Parameters for the first
with the observed GN (r) as depicted in Figures 3a and 3b.
Final results of the least squares fit are summarized in Tables 3
and 4. For the sample involving NaCl, the present value of
the nearest neighbor N£Ow distance, rNO = 3.09 ¡ is slightly
longer than that reported for more concentrated aqueous
15 mol % urea solutions by means of the neutron diffrac-
nearest neighbor N£D O interaction, rNO, lNO, lND, and nNO
,
2
were treated as independent parameters. b) The tilt angle ¡
describing angle between the N£OW axis and the molecular
1
4
15
4
plane of D O in the first hydration shell of the amino group was
tion with N/ N isotopic substitution (r = 2.92(1) ¡). The
NO
2

1
08
Bull. Chem. Soc. Jpn. Vol. 86, No. 1 (2013)
Structure of Concentrated Urea Solutions
Table 4. Results of the Least-Squares Refinement for
inter
¦
N
(Q) Observed for Aqueous 10 mol % Urea Solutions
Involving 10 mol % LiCl in D2O
Interaction
N£D2O(I)
i£j
rij/¡
lij/¡
nij
N£O
N£D
= 66(1)°
3.19(1)
0.17(1)
0.27(1)
3.3(1)
(6.6)
a)
b)
(3.56)
c)
¡
N£D2O(II)
N£D2O(III)
N£D2O
N£D2O
3.93(1)
5.44(1)
0.45(1)
1.22(1)
4.1(1)
19.3(2)
r0/¡
l0/¡
N£Xd)
5.90(2)
1.36(1)
a) Calculated from the values rNO = 3.19 ¡ and ¡ = 66°.
b) 2nNO. c) The angle between N£O axis and the molecular
plane of D2O. d) X: Li, Cl, O, D, N, and C.
larger than the values of aqueous 15 mol % urea solutions.4
The present results indicate that the first hydration shell of
amino group of the urea molecule is significantly modified by
coexisting ions. Although the longer nearest neighbor N£OW
distance suggests weakened interaction between the amino
group and the nearest neighbor water molecules, the number of
water molecules in the first hydration shell is rather increased
compared with that reported for aqueous urea solutions. It can
be suggested that the first hydration shell of amino group of
urea molecule has considerable flexibility. This may explain
exceptionally higher solubility of urea in aqueous solutions
involving NaCl and LiCl. It is supposed that one water
molecule is simultaneously shared by a urea molecule and
coexisting ions. Details of hydration structure of coexisting
ions are discussed in the following section.
Figure 3. Observed intermolecular distribution function
inter
around the nitrogen atom of the urea molecule, GN (r)
(
(
dots), and Fourier transform of the calculated ¦Ncalc(Q)
solid line), for aqueous 10 mol % urea solutions involving
5
mol % NaCl (a) and 10 mol % LiCl (b), respectively. The
contributions from the short- and long-range interactions
are denoted by broken lines.
¹
+
Hydration Structure around Cl and Li . The observed
difference functions, ¦ (Q) and ¦ (Q), are represented in
Cl
Li
Table 3. Results of the Least-Squares Refinement for
Figures 4a and 4c (dots), respectively. Although data points are
somewhat scattered due to statistical uncertainties, oscillational
features are evidently observed in the present difference
functions. Interference features observed for the present
inter
¦
N
(Q) Observed for Aqueous 10 mol % Urea Solutions
Involving 5 mol % NaCl in D2O
Interaction
N£D2O(I)
i£j
rij/¡
lij/¡
nij
¦ (Q) are very similar to those reported for various aqueous
Cl
N£O
N£D
= 54(1)°
3.09(1) 0.14(1)
2.7(1)
(5.4)
1
4,26,27
a)
b)
(3.56)
0.26(1)
solutions,
in which water molecules within the first
c)
¹
¡
hydration shell face one hydrogen atom toward Cl . The dis-
tribution function around Cl , G (r), observed for urea­
¹
Cl
N£D2O(II)
N£D2O(III)
N£D2O
N£D2O
4.02(1) 0.69(1)
5.04(1) 1.20(1)
3.8(1)
17.3(2)
Na*Cl­D O solution (Figure 5a) exhibits the first peak located
2
at r = 2.26 ¡ which is attributable to the nearest neighbor
¹
14,26,27
Cl £DW1(­OW­DW2) interaction.
The sum of contribu-
¹
¹
r0/¡
l0/¡
tions from the nearest neighbor Cl £O and Cl £D inter-
W
W2
N£Xd)
5.61(1) 1.07(1)
actions should be involved in the second peak at r = 3.8 ¡
in the present G (r).
Cl
a) Calculated from the values rNO = 3.09 ¡ and ¡ = 54°.
b) 2nNO. c) The angle between N£O axis and the molecular
plane of D2O. d) X: Na, Cl, O, D, N, and C.
Structural parameters concerning the first hydration shell of
¹
Cl were determined through the least-squares fitting of the
observed ¦ (Q). The following assumptions were adopted in
Cl
evaluating the model function. a) Parameters for the nearest
¹
¹
¹
number of water molecules hydrogen bonded with the amino
group of urea was obtained to be 2.7(1), which is somewhat
neighbor Cl £D2O interaction, r(Cl £DW1), l(Cl £DW1),
¹
and n(Cl £D ), were treated as independent parameters.
W1
¹
larger than the value 2.0(1) obtained from aqueous 15 mol %
b) The tilt angle ¡, defined by ¾Cl £D ­O , was allow
to vary independently. Interatomic distances, r(Cl £O ) and
r(Cl £DW2), were calculated using known molecular geometry
W1
W
4
¹
urea solutions. Present values, rNO = 3.19(1) ¡ and nNO
=
W
¹
3
.3(1), determined for sample solutions involving LiCl are also

T. Miyazaki et al.
Bull. Chem. Soc. Jpn. Vol. 86, No. 1 (2013)
109
Figure 4. a) ¦Cl(Q), observed for aqueous 10 mol % urea
solutions involving 5 mol % NaCl (dots), and the best-fit of
the calculated ¦Cl (Q) (solid line). c) ¦Li(Q), observed
¹
Figure 5. a) Observed distribution function around Cl ,
calc
G (r) (dots), and Fourier transform of the calculated
Cl (Q) (solid line), for aqueous 10 mol % urea solutions
involving 5 mol % NaCl. b) Observed distribution function
around Li , GLi(r) (dots), and Fourier transform of the
Cl
calc
¦
for aqueous 10 mol % urea solutions involving 10 mol %
calc
LiCl (dots), and the best-fit of the calculated ¦Li (Q)
+
(solid line). b), d) Residual functions (dots).
calc
calculated ¦Li (Q) (solid line), for aqueous 10 mol %
urea solutions involving 10 mol % LiCl. The contributions
from the short- and long-range interactions are denoted by
broken lines.
Table 5. Results of the Least-Squares Refinement for
Cl(Q) Observed for Aqueous 10 mol % Urea Solutions
Involving 5 mol % NaCl in D2O
¦
Interaction
i£j
rij/¡
lij/¡
nij
¹
¹
of D2O was introduced, however, statistical uncertainties
Cl £D2O(I) Cl £D1
2.27(1)
0.24(1)
5.8(3)
¹
a)
b)
c)
involved in the observed ¦ (Q) prevented to determine the
Cl £O
(3.16)
(3.83)
(0.28)
(0.31)
(5.8)
(5.8)
Cl
¹
a)
b)
c)
optimized value from the least-squares fitting procedure.
Therefore, value of ¢ was tentatively fixed to 0° during the
Cl £D2
c)
d)
¡
= 150(4)°
(¢ = 0°)
present least-squares analysis. d) The contribution from the
¹
¹
¹
Cl £D2O(II) Cl £D2O
3.66(3)
0.58(8)
5.5(5)
second nearest neighbor Cl £D O(II) interaction was involved
2
in the model function in order to improve the fit in the low-Q
region. This contribution was treated as single interaction with
the coherent scattering length b¡ being 2bD + bO in eq 10.
e) Long-range parameters, r and l , were allowed to vary
r0/¡
l0/¡
¹
e)
Cl £X
4.30(4)
0.43(6)
0
0
a) Calculated from the values rClD = 2.27 ¡, ¡ = 150°, and
1/2
independently. The fitting procedure was performed by the
¢
= 0°. b) lCl¡ = lClD(rCl¡/rClD) . c) Fixed at the value of
2
3
SALS program.
nClD1. d) Fixed. e) X: Na, Cl, O, D, N, and C.
The best-fit result is shown in Figure 4a. Interference
features in the observed ¦ (Q) are satisfactorily reproduced
Cl
2
4,25
of D O in the liquid state.
The root-mean-square ampli-
in the entire Q-range (Figure 4b). The final results of inde-
2
¹
¹
¹
tudes, l(Cl £O ) and l(Cl £D ), were evaluated from the
pendent parameters are listed in Table 5. The Cl £D
W
W2
W1
¹
¹
20
values of l(Cl £DW1) and r(Cl £DW1) as follows,
distance, 2.27(1) ¡, is in good agreement with that reported
¹
by previous neutron diffraction studies, r(Cl £DW1) = 2.22­
ꢀ
ꢀ
ꢀ
ꢀ
1=2
lðCl £ZÞ ¼ lðCl £DW1Þ ꢂ ½rðCl £ZÞ=rðCl £DW1Þꢁ
ð12Þ
2
6,27
¹
2
.29 ¡.
The present value of n(Cl £DW1) = 5.8(2) also
where, Z stands for O or DW2. c) A dihedral angle ¢ between
plane involving atoms, Cl £DW1­OW and molecular plane
agrees well with that reported for various aqueous solutions
in which the contact ion pair formation with cation does not
W
¹

110
Bull. Chem. Soc. Jpn. Vol. 86, No. 1 (2013)
Structure of Concentrated Urea Solutions
Table 6. Results of the Least-Squares Refinement for ¦Li(Q)
Observed for Aqueous 10 mol % Urea Solutions Involving
urea. If we assume the number of D O molecules in the first
hydration shell of Cl is 5.8, the same value as that derived
2
¹
10 mol % LiCl in D2O
from the present urea­NaCl­D O solution, the sum of the
2
number of D O molecules in the first hydration shell of amino
2
Interaction
i£j
rij/¡
lij/¡
nij
+
¹
group of urea, Li , and Cl , exceed twice the total stoichio-
metric number of D O. This implies that each D O molecule is
+
+
Li £D2O(I)
Li £O
1.95(1)
2.64(1)
0.09(1)
0.33(1)
5.2(1)
(10.4)
2
2
+
a)
Li £D
simultaneously shared by at least two ions (or urea molecule).
In order to obtain more detailed information on the intermo-
+
+
Li £D2O(II)
Li £D2O
4.25(1)
0.44(1)
7.2(1)
lecular structure of the urea­MCl­D O solutions, it is necessary
2
+
¹
to deduce partial structure functions, N£M , N£Cl , and
M £Cl , which requires additional neutron diffraction mea-
surements. This will be a future subject.
r0/¡
l0/¡
+
¹
+
b)
Li £X
4.41(1)
0.58(1)
a) 2nLiO. b) X: Li, Cl, O, D, N, and C.
We thank the Institute of Solid State Physics (ISSP), the
University of Tokyo, for allowing us to use the 4G (GPTAS)
diffractometer. We are grateful to Prof. Taku J. Sato (ISSP, the
University of Tokyo) for his help during the course of neutron
diffraction measurements on the 4G diffractometer.
occur, n(Cl £D ) = 5.5­5.9.26,27 In the present solution, Cl
¹
¹
W1
is considered to have stable first hydration shell involving ca. 6
water molecules.
Difference function, ¦ (Q), observed for the present urea­
Li
*
LiCl­D O solution exhibits well defined oscillatory features
2
References
as shown in Figure 4c. Fourier transform of the observed
¦
+
(Q) gives the distribution function around Li , G (r), as
1
27.
2
J. L. Finney, J. Turner, Ann. N.Y. Acad. Sci. 1986, 482,
Li
Li
1
depicted in Figure 5b. The short-range peaks appearing in
G (r) represent the nearest neighbor Li £O and Li £D
W
interactions, which have frequently been reported in G (r) for
various aqueous solutions.
hydration shell of Li can appear as the third peak at r = 4.4 ¡
in the present G (r). In order to obtain parameters concerning
the short-range structure around Li , least-squares fitting was
adopted to the observed ¦ (Q). The following assumptions
were employed in evaluation of theoretical ¦Li
tion. a) Structural parameters for the nearest neighbor Li £D O
interaction, rLiO, lLiO, nLiO, rLiD, and lLiD, were treated as
independent parameters. The value of nLiD was fixed to 2nLiO
during the fitting procedure. b) Contribution from the second
+
+
J. L. Finney, J. Turner, Electrochim. Acta 1988, 33, 1183.
J. Turner, J. L. Finney, A. K. Soper, Z. Naturforsch., A
Li
W
3
Li
28­32
1991, 46, 73.
An indication of the second
+
4
Y. Kameda, H. Naganuma, K. Mochiduki, M. Imano, T.
Usuki, O. Uemura, Bull. Chem. Soc. Jpn. 2002, 75, 2579.
Li
5
Y. Kameda, A. Maki, Y. Amo, T. Usuki, Bull. Chem. Soc.
+
Jpn. 2010, 83, 131.
Li
6
A. K. Soper, E. W. Castner, Jr., A. Luzar, Biophys. Chem.
model
(Q) func-
2
003, 105, 649.
+
2
7
R. C. Burton, E. S. Ferrari, R. J. Davey, J. Hopwood, M. J.
Quayle, J. L. Finney, D. T. Bowron, Cryst. Growth Des. 2008, 8,
1559.
8
J. L. Finney, A. K. Soper, J. Turner, Physica B 1989, 156­
+
hydration shell of Li was taken into account. The interaction
1
57, 151.
was treated as single interaction with the coherent scattering
length b = 2b + b . c) Long-range parameters, r and l ,
9
Y. Kameda, M. Sasaki, S. Hino, Y. Amo, T. Usuki, Bull.
¡
D
O
0
0
Chem. Soc. Jpn. 2006, 79, 1367.
were allowed to vary independently. The fitting procedure was
carried out by the SALS program.23
10 H. H. Paalman, C. J. Pings, J. Appl. Phys. 1962, 33,
2635.
The best-fit results are compared with the observed ¦ (Q)
11 I. A. Blech, B. L. Averbach, Phys. Rev. 1965, 137, A1113.
Li
1
384.
2
Y. Kameda, O. Uemura, Bull. Chem. Soc. Jpn. 1993, 66,
in Figure 4c. Residual functions are indicated in Figure 4d.
model
Fourier transform of calculated ¦Li
(Q) well reproduce
1
3
Y. Kameda, T. Usuki, O. Uemura, Bull. Chem. Soc. Jpn.
observed G (r) as depicted in Figure 5b. Final values of
Li
1
998, 71, 1305.
independent parameters are summarized in Table 6. The
+
14 A. K. Soper, G. W. Neilson, J. E. Endeby, R. A. Howe,
present value, r(Li £O ) = 1.95(1) ¡, is in excellent agree-
W
1
2,28,30
J. Phys. C: Solid State Phys. 1977, 10, 1793.
J. E. Enderby, G. W. Neilson, in Water: A Comprehensive
ment with that reported for various aqueous solutions.
1
5
The number of water molecules in the first hydration shell,
Treatise, ed. by F. Franks, Plenum Press, New York, 1979, Vol. 6,
pp. 1­46. doi:10.1007/978-1-4684-8018-4_1.
+
n(Li £O ), was determined to 5.2(1), which agrees well with
W
2
8
30
values, 5.5(3), and 6.0(4), observed for 3.6 m (6.7 mol %)
LiCl­D O solutions. The first hydration shell of Li seems well
1
1
6
7
J. E. Enderby, Chem. Soc. Rev. 1995, 24, 159.
S. Swaminathan, B. M. Craven, R. K. McMullan, Acta
+
2
maintained regardless of coexistence of urea molecules. The
Crystallogr., Sect. B 1984, 40, 300.
1
257.
1
tilt angle of D O molecule, defined by the angle between the
2
8
K. Iijima, K. Tanaka, S. Onuma, J. Mol. Struct. 1991, 246,
+
Li £O axis and molecular plane of D O, was estimated to be
8(15)° from observed values for r(Li £OW), r(Li £DW), and
known molecular geometry of D O in the liquid phase.
W
2
+
+
1
9
C. Kato, S. Konaka, T. Iijima, M. Kimura, Bull. Chem. Soc.
2
4,25
The
2
Jpn. 1969, 42, 2148.
present value of the tilt angle is smaller than the typical value
20 A. H. Narten, M. D. Danford, H. A. Levy, Discuss.
Faraday Soc. 1967, 43, 97.
21 R. Caminiti, P. Cucca, M. Monduzzi, G. Saba, G. Crisponi,
2
8
(
40(5)°), suggesting that the orientation of D O molecules in
2
+
the first hydration shell of Li can be affected by neighboring

T. Miyazaki et al.
Bull. Chem. Soc. Jpn. Vol. 86, No. 1 (2013)
111
J. Chem. Phys. 1984, 81, 543.
P. S. Salmon, Faraday Discuss. Chem. Soc. 1988, 85, 137.
2
2
2
3
H. Ohtaki, N. Fukushima, J. Solution Chem. 1992, 21, 23.
T. Nakagawa, Y. Oyanagi, Recent Developments in Stat-
28 J. R. Newsome, G. W. Neilson, J. E. Enderby, J. Phys. C:
Solid State Phys. 1980, 13, L923.
istical Inference and Data Analysis, North-Holland, Amsterdam,
29 M. Yamagami, T. Yamaguchi, H. Wakita, M. Misawa,
J. Chem. Phys. 1994, 100, 3122.
1
980, p. 221.
2
2
4
5
J. G. Powles, Mol. Phys. 1981, 42, 757.
Y. Kameda, O. Uemura, Bull. Chem. Soc. Jpn. 1992, 65,
30 I. Howell, G. W. Neilson, J. Phys.: Condens. Matter 1996,
8, 4455.
2
021.
31 S. Ansell, J. Dupuy-Philon, J.-F. Jal, G. W. Neilson,
J. Phys.: Condens. Matter 1997, 9, 8835.
2
6
J. E. Enderby, S. Cummings, G. J. Herdman, G. W. Neilson,
P. S. Salmon, N. Skipper, J. Phys. Chem. 1987, 91, 5851.
D. H. Powell, A. C. Barnes, J. E. Ederby, G. W. Neilson,
32 Y. Kameda, K. Mochiduki, M. Imano, H. Naganuma, M.
Sasaki, Y. Amo, T. Usuki, J. Mol. Liq. 2005, 119, 159.
2
7