Article

pubs.acs.org/jced

Phase Equilibria in the Urea−Biuret−Water System

Alexey L. Voskov,† Tatyana S. Babkina,† Alexander V. Kuznetsov,‡ and Irina A. Uspenskaya*,†
†
Chemistry Department, Lomonosov Moscow State University, 119991 Moscow, Russia
‡
Department of Materials Science, Lomonosov Moscow State University, 119991 Moscow, Russia

ABSTRACT: A thermodynamic assessment of the urea−biuret−water

ternary system and its binary subsystems has been made. The Margules
model has been applied for the description of excess Gibbs energies of binary
solutions, and the properties of ternary solutions have been assessed by using
the Muggiani method. Thermodynamic properties of
(NH2CO)2NH·(NH2)2CO (I) and (NH2CO)2NH·0.7H2O (II) have been
estimated from the condition of equilibrium between the solution and
stochiometric phases. The liquidus surface and isothermal sections of the
phase diagram of the urea−biuret−water system at the temperature interval
T = (268 to 373) K have been calculated. The comparison of calculated and experimental data is discussed.

■ INTRODUCTION
Urea is one of the most important and demanded nitrogen
fertilizers. It manufactured by means of the Bazarov reaction,
that is, interaction between carbon dioxide and ammonia at
high pressures and temperatures. One of the possible
byproducts is biuret, and its formation depends on the
conditions of the process. This substance has a negative effect
on plant growth, and the upper limit of its fraction in the final
product should not be larger than 3 wt %. A knowledge of the
liquidus in the urea−biuret−water ternary system is of practical
interest, but there are no systematic studies of phase equilibria
in this system.
The water−urea system is the most investigated of the three
binary subsystems, and the solubility of (NH2)2CO was
measured by Speyers,1 Pinck and Kelly,2 Janecke,3 Shnidman
and Sunier,4 Chadwell and Politi,5 Kakinuma,6 Frejacques,7
Miller and Dittmar,8 and Babkina and Kuznetsov9 using
different methods. Phase diagrams of biuret−water and
biuret−urea systems were published in the articles of Rollet
et al.10,11 and Babkina and Kuznetsov.9 Isothermal sections of
the ternary phase diagram are shown in the monography of
Kucheriavij and Lebedev.12 In the work of Barone et al.13 the
enthalpy of dilution of urea and biuret aqueous solutions as well
as the enthalpy of mixing biuret and urea water solutions by the
flow microcalorimetry method at T = 298 K was investigated.
Infinitely diluted water solutions of components were used as
their standard states. It was shown that obtained data can be
adequately described by using one term in the excess Gibbs
energy of solution; that is, the solutions are “regular”.
■ THERMODYNAMIC MODELS OF PHASES IN
BINARY AND TERNARY SYSTEMS
There are liquid solutions and stoichiometric solid phases in the
biuret−urea−water ternary system. All components are
insoluble in the solid state. Only two compounds,
(NH2CO)2NH·(NH2)2CO (I) and (NH2CO)2NH·0.7H2O
(II), are present in the binary subsystems, and no information
about ternary solid phases have been reported.
Model of Liquid. The liquid phase was described using the
Margules−Muggiani formalism. The Gibbs energy of an A−B
binary solution was approximated by the Margules equation:
Δmix G(T , x) = −TS id(x) + Δmix Gex (x)
= RT ((1 − x)ln(1 − x) + x ln x)
+ x(1 − x)[(1 − x)ABA + xAAB] (1)
The corresponding expressions for the chemical potentials of
components A and B are:
μAl = μA° ,l + RT ln(1 − x) + x 2[AAB + 2(ABA − AAB)
(1 − x)] (2)
μBl = μB° ,l + RT ln x + (1 − x)2 [ABA + 2(AAB − ABA )x]
(3)
where μ°i is the Gibbs energy of pure liquid i; that is, the
,l

The aim of this work is the thermodynamic assessment of the symmetric system was chosen to represent the thermodynamic
(NH2)2CO−(NH2CO)2NH−H2O ternary system based on the functions of melts. The Gibbs energies of melting were
existing experimental data and calculation of isothermal estimated using eq 4:
sections of the ternary phase diagram. The convex hull
methods have been implemented in two packages PhDi and Received: July 25, 2012
TernAPI; the latter one was used in the present study for the Accepted: August 30, 2012
calculation of phase diagrams.14,15 Published: September 26, 2012

© 2012 American Chemical Society 3225 dx.doi.org/10.1021/je300832f | J. Chem. Eng. Data 2012, 57, 3225−3232
Journal of Chemical & Engineering Data Article

Δ m G ( T ) = Δ m H (T ) − T Δ m S (T ) Cpbiuret,l(Tm) Cpurea,l(Tm)
=
= Δm H(Tm) + Δm Cp(T − Tm) Cpbiuret,s(Tm) Cpurea,s(Tm)
⎛ Δ H(Tm) T ⎞ ⇔ Δm Cpbiuret
− T⎜ m + Δm Cp ln ⎟
⎝ Tm Tm ⎠ Cpbiuret,s(Tm)
= Δm Cpurea
= (Δm H(Tm) − Δm CpTm) Cpurea,s(Tm)
⎛ Δ H(Tm) ⎞ ⎛ C urea,l ⎞
+ T ⎜Δm Cp(1 + ln Tm) − m ⎟ = Cpbiuret,s(Tm)⎜⎜ urea,s − 1⎟⎟
p
⎝ Tm ⎠
⎝ Cp ⎠ (8)
− Δm CpT ln T (4)
The model parameters AAB and ABA in eq 1 were evaluated
In this equation it was assumed that the ΔmCp = const. This using MATLAB software, based on the experimental liquidus
assumption is reasonable due to a relatively narrow temperature data available in the literature.
range under investigation. Three constant parameters for each The Muggiani method22 was used to obtain the Gibbs energy
component are required in this case: the temperature of fusion, of the ternary liquid from the Gibbs energies of binary liquids:
Tm, enthalpy of fusion at Tm, ΔmH(Tm), and change of heat 3
capacity during melting, ΔmCp. Their values for biuret, urea, Δmix G = RT ∑ xi ln xi + Δmix Gex
and water are listed in Table 1. i=1 (9)

Table 1. Thermodynamic Properties of Components (Urea, 4x 2x3

Δmix Gex = ex
[Δmix G23 ]x2′
Biuret, and Water) That Are Required for eq 4 (2x 2 + x1)(2x3 + x1)
ΔmH(Tm) Tm ΔmCp 4x1x3 ex
+ [Δmix G13 ]x1′
component J·mol−1 K J·mol−1·K−1 (2x1 + x 2)(2x3 + x 2)
urea 14644 ± 500 405.85 ± 0.5 33.95 ± 8a 4x1x 2 ex
biuret 26100 ± 500 473.8 ± 0.4 55.98 ± 13a + [Δmix G12 ]x1″
(2x1 + x3)(2x 2 + x3) (10)
water 6010 ± 4 273.16 ± 0.01 38.21 ± 0.05
a
Estimated in this work. where numbers 1, 2, and 3 correspond to the urea, biuret, and
water; x2′ = x2 + x1/2, x1′ = x1 + x2/2, x1″ = x1 + x3/2. The values
of ΔmixGexij in square brackets were calculated by eq 1 with the
corresponding set of parameters AAB and ABA for each
In this work the values Tm and ΔmH(Tm) for urea and biuret subsystem. No ternary parameters were used for the liquid
recommended in ref 9 were used. They were obtained as the phase.
Stoichiometric Phases. Both stoichiometric phases,
mean of the most reliable literature data. The values of Tm,
(NH2CO)2NH·(NH2)2CO and (NH2CO)2NH·0.7H2O, melt
ΔmH(Tm), and ΔmCp for water were taken from ref 16. incongruently at (389.8 and 384.1) K, respectively.9 There are
The experimental heat capacity of solid urea17−19 in the no experimentally obtained values of thermodynamic properties
temperature interval (240 to 400) K was described by the of phases I and II except for the heat of solution of biuret
truncated equation of Berman and Brown20 (Cp = a + bT−0.5, hydrate.23 Therefore, the Gibbs energies of these binary
compounds at various temperatures were estimated by using
where b < 0):
the tangent line method24 (see Figure 1).
Cpurea,s(T ) = (253.65 ± 4) − (2763.3 ± 67)T −0.5 J· mol−1·K−1
Consider a binary solid compound in equilibrium with liquid.
According to the tangent line method, the Gibbs energy of the
(5) reaction
The heat capacity of liquid urea near the melting point was (1 − x s)A(liquid) + x s B(liquid) = A1 − xBx (solid) (11)
taken from ref 18 as the average of seven experimental points.
which corresponds to formation of a binary solid compound
−1 −1 from pure liquid components and can be expressed as
Cpliq(Tm) = (150.43 ± 7) J·mol ·K (6)
∂Δmix G(T )
Equation 7 from ref 21 for the calculation of the heat capacity Δf G(T ) = Δmix G(T )|x = x l + (x s − x l)
∂x x=xl
of solid biuret over the temperature interval from (240 to 450)
∂Δmix G(T )
K has been used: = μ1liq (x l , T ) + x s
∂x x = xl (12)
Cp(T ) = (27.04 ± 1) + (0.3483 ± 0.0030)T J·mol−1·K−1
(7) ∂Δmix G(T )
= RT (ln x − ln(1 − x)) + (1 − 2x)
∂x
The difference between the heat capacity of liquid and solid
((1 − x)ABA + AAB) + x(1 − x)
biuret was estimated assuming equality of the relative heat
capacity change during melting for biuret and urea: (AAB − ABA ) (13)

3226 dx.doi.org/10.1021/je300832f | J. Chem. Eng. Data 2012, 57, 3225−3232

Journal of Chemical & Engineering Data Article

of data points and the values of the target function are

listed in Table 2.

Table 2. Some Statistic Characteristics of Thermodynamic

Models of Melts in the Binary System with Urea, Biuret, and
Water (N is the Number of Points, σ is a Value of Target
Function)
system N σ
urea−biuret 17 1.3·10−3
water−biuret 4 9.6·10−5
water−urea 80 3.0·10−2

2. Values of ΔfH298 and ΔfS298 for the binary stoichiometric

compounds were obtained by the minimization of the
target function:
⎛ T calc − T exp ⎞2 ⎛ T calc − T exp ⎞2
⎟⎟ ωeut + ⎜⎜ ⎟ ωper
per per
σ = ⎜⎜ eut
exp
eut 2
exp ⎟
2
Figure 1. Tangent line connecting Gibbs energies of coexisting liquid ⎝ Teut ⎠ ⎝ T per ⎠
and solid phases in the water−biuret system at T = 380 K. The curve is
the Gibbs energy of liquid; the solid line is a tangent line, and symbols ⎛ Δ G fit − Δ G calc ⎞2
and dashed lines indicate values of variables (the molar fraction of + ∑ ⎜⎜ f i f i
calc
⎟⎟ (ωiG)2
biuret and Gibbs energy) of coexisting liquid (l) and stoichiometric (s) i ⎝ Δf Gi ⎠ (17)
phases.
where superscripts calc and exp correspond to calculated
l s
where x and x are the compositions of the liquid and solid and experimental values, respectively, ΔfGcalc
i is calculated
phases, respectively. The values of ΔfG(T) calculated at various from eqs 12 and 13, and ΔfGfit i values are obtained from
temperatures can be fitted to obtain ΔfH298 and ΔfS298 the eq 14. Teut and Tper are the eutectic and peritectic
assuming that the enthalpy and entropy of reaction 11 are temperatures. In the case of the urea−biuret system, ωeut
independent of temperature. = 1 and ωper = 2.5, and in the system water−biuret, ωper
= 0. In all cases, ωGi = 1. Thermodynamic functions
Δf G(T ) = Δf H298 − T Δf S298 (14) defined this way are essentially fictitious values because
To calculate Gibbs energy formation from pure solid they are obtained by solving mathematically incorrect
components the Gibbs energy of melting of pure substances inverse task of thermodynamics (an inverse task is a
are necessary: reconstruction of thermodynamic properties from
equilibria only, this task has non single decision unlike
Δf G°(T ) = Δf G(T ) + (1 − x s)Δm GA (T ) + x sΔm G B(T ) a direct task, where the equilibrium condition is
(15) calculated on the basis of known thermodynamic

■
properties of coexisting phases). In ref 25 it was shown
OPTIMIZATION OF MODEL PARAMETERS that it is possible to reproduce conditions of phase
The parameters of the Margules model for the liquid phase and equilibria by means of these values with high accuracy,
the ΔfH298 and ΔfS298 values for each binary compound were but the probability of “guessing” real value of
obtained by a nonlinear least-squares method, using exper- thermodynamic property is low.
imental data on phase equilibria. The optimization was carried To find the confidence intervals of the model parameters, the
out in two steps: Jacobian
1. The model parameters for the liquid phase were obtained
from the liquidus in the binary systems by minimization ⎛ ∂F ⎞
of the following (target function): Jij = ⎜⎜ i ⎟⎟
⎝ ∂θj ⎠θ = θ
⎡ m 0, m (18)
⎛ T exp − T calc ⎞2
⎢
σ = ∑ (xi − xi ) (ωi ) + ⎜⎜ i exp i ⎟⎟
exp calc 2 (x) 2
was calculated. Here, 1 ≤ i ≤ n, 1 ≤ j ≤ m, n is the total number
i ⎣
⎢ ⎝ Ti ⎠ of experimental points, m is the number of parameters to be
⎤ estimated (i.e., nonlinear regression coefficients) Fi(θ1, ..., θp) is
(T ) 2 ⎥ the i-th component of the vector function F (it is a difference
(ωi ) between predicted and experimental values), θj denotes model
⎥
⎦ (16) parameters, and θ0,m stands for optimized model parameters.
exp exp calc calc Target functions (eq 16 and 17) can be represented as:
where (xi ,Ti ) and (xi ,Ti ) are experimental and
calculated coordinates of the ith liquidus point,
respectively, and ω(x) and ω(T) are the corresponding
σ= ∑ Fi2
i i (19)
i
statistical weights. In the water−biuret and urea−biuret
−1
binary systems, ω(x) i = 1 and ω(T) i = 0. For the water− We calculated the covariance matrix C = (J J) , which T

urea system, ωi = 1 and ω(T)

(x)
i = 0 for ordinary liquidus diagonal elements Cjj were used to calculate the confidence
points but for the eutectic, ω(x) i = ωi
(T)
= 20. The number intervals of model parameters as
3227 dx.doi.org/10.1021/je300832f | J. Chem. Eng. Data 2012, 57, 3225−3232
Journal of Chemical & Engineering Data Article

Δθj = t0.05, v CjjsR2 (20)

Δf G(0.5[(NH 2CO)2 NH ·(NH 2)2 CO])

where t0.05,v is the Student test value for v = n − m degrees of = −(23.509 ± 10) ·103 + (53.416 ± 30)T J·mol−1·K−1
freedom and a 95 % probability and s2R is the residual variance. (21)

■ RESULTS AND DISCUSSION

Binary Systems. The first study of the phase diagram of the
Δf G°(0.5[(NH 2CO)2 NH ·(NH 2)2 CO])
= −(5.852 ± 10) ·103 + (14.214 ± 30)T J·mol−1·K−1
urea−biuret was reported by Rollet et al.10 The components (22)
were mixed in various ratios over the whole composition range Phase equilibria were calculated assuming negligible mutual
and studied by the thermal analysis method. During the solubility of solid components and so-called regular inter-
experiments, an interesting phenomenon was observed: first particle interactions in the liquid phase. Parameters of eq 1 for
melting curves at random had a plateau at (379.1 or 384.1) K, the urea−biuret subsystem are listed in Table 3. The results of
and heating conditions did not affect its position contrary
unlike to the second round of melting. Slow cooling of melts Table 3. Parameters ABA and AAB of the Liquid Model Given
resulted in a plateau at 384.1 K, while rapid cooling produced a by eq 1 for the Binary Subsystems of the Urea(1)−
plateau at 379.1 K. The authors attributed these observations to Biuret(2)−Water(3) Ternary System
the removal of ammonia at temperatures above 400 K, which
would result in increasing the biuret content in the system. subsystem ABA/J·mol−1 AAB/J·mol−1
However, we believe that the true reason for this phenomenon (1)−(2) −(372 ± 169) −(372 ± 169)
is the formation of a stable or metastable eutectic depending on (2)−(3) (466 ± 198) (466 ± 198)
the cooling and heating conditions. If to kinetic restrictions the (3)−(1) (648 ± 63) −(390 ± 196)
stoichiometric phase with equimolar ratio of components is not
formed, the metastable eutectic is observed at 379.1 K. If there Rollet et al.10 at temperatures higher than 420 K were not used
is enough time to form the compound, the stable eutectic point for the determination of these parameters due to possible
is detected at 384.1 K. As can be seen from Figure 2, our decomposition of biuret during the experiments. The thermal
stability of biuret and the influence of the heating rate on the
melting point of (NH2CO)2NH were discussed in detail
earlier.9 Invariant points of the urea−biuret phase diagram are
given in Table 4. As can be seen from Figure 2, a good

Table 4. Calculated Coordinates of Invariant Points in the

Urea(1)−Biuret(2)−Water(3) Ternary Systema
special point T/K x2 x3
(1)−(2) eutectic 384.0 0.2095 0.0000
(1)−(2) peritectic 390.3 0.2937 0.0000
(1)−(2) eutectic 379.6 0.2469 0.0000
(no compounds)b
(2)−(3) eutectic 273.1 0.0009 0.9991
(2)−(3) peritectic 384.4 0.2300 0.7700
(1)−(3) eutectic 260.6 0.0000 0.8763
(1)−(2)−(3) eutectic 260.6 0.0004 0.8760
(1)−(2)−(3) eutectic 259.1 0.0170 0.8616
(no compounds)b
(1)−(2)−(3) peritecticc 363.4 0.1723 0.4268
Figure 2. Phase diagram of the (NH2)2CO−(NH2CO)2NH binary a
system. Solid lines indicate the calculated diagram, dashed lines the Standard uncertainties u are: u(T) = 0.1 K and u(x) = 0.0001. These
metastable eutectic equilibria, and symbols the experimental data: ○, uncertainties are not deviations of calculated values from the
Rollet et al.;10 ▽, Babkina and Kuznetsov.9 experiment. bMetastable equilibria without binary compounds.
c
Coexistence of liquid phase, solid biuret, and binary compounds I
and II.

calculations substantiated this hypothesis (see Figure 2). The
solid lines in Figure 2 correspond to the stable phase diagram
while the dashed show metastable phase equilibria, symbols
illustrate experimental data.
Thermodynamic functions of the stoichiometric phase
(NH2CO)2NH·(NH2)2CO were estimated as described in
previous sections. The composition of this phase reported in
the work of Babkina and Kuznetsov9 was accepted in the
present study. The temperature dependence of the Gibbs
energy of formation from liquid (eq 21) or from solid
components (eq 22) can be expressed as
3228
agreement was obtained between the calculated lines and the
experimental points. This is to be expected, since most of the
experimental data were used for the assessment of model
parameters.
The phase diagram of the water−biuret system was studied
in ref 11 at normal and elevated pressures. The equilibrium
phase boundaries were obtained by thermal analysis (analysis of
heating and cooling curves) and by determination of the
solubility limits, using visual observation of the disappearance
of the solid phases. A good agreement between the results
obtained by different methods was reported by Rollet and
Cohen-Adad.11 The solubility of biuret in water at 273 K is
dx.doi.org/10.1021/je300832f | J. Chem. Eng. Data 2012, 57, 3225−3232
Journal of Chemical & Engineering Data Article

negligible and increases rapidly with increasing temperature,

reaching 1.2 wt %, 7.0 wt %, and 20.0 wt % at (298, 323, and
348) K, respectively. At the normal boiling point of the
saturated solution (378.2 K), the content of biuret is equal to
53.5 wt %. At atmospheric pressure crystalline monohydrate of
biuret is in the equilibrium with the liquid phase. Some
refinements of the phase diagram and the composition of the
crystal hydrate were reported earlier.9 The correct formula of
the hydrate was determined to be C2N3O2H5·(0.70 ±
0.05)H2O.
A good description of the phase boundaries in the binary
water−biuret system was obtained with a single parameter AAB
= ABA in eq 1 (see the second line in Table 3). The estimated
temperature dependence of the Gibbs energy of formation from
liquid components (eq 23) and from liquid water and solid
biuret (eq 24) may be expressed as
⎛ 1 ⎞
Δf G⎜ [(NH 2CO)2 NH· 0.7H 2O]⎟
⎝ 1.7 ⎠
Figure 3. Phase diagram of the H2O−(NH2CO)2NH binary system.
= −(24.273 ± 2) ·103 + (55.515 ± 5)T J·mol−1·K−1 Solid lines indicate the calculated diagram, symbols the experimental
(23) data: ○, Rollet and Cohen-Adad;11 ▽, Babkina and Kuznetsov.9

⎛ 1 ⎞
Δf G◦⎜ [(NH 2CO)2 NH· 0.7H 2O]⎟
⎝ 1.7 ⎠
= −(13.682 ± 2) ·103 + (34.932 ± 5)T J·mol−1·K−1
(24)
Experimental values of the heat released during dissolution of
solid biuret and hydrate of biuret in water were published in ref
23. The recalculation of the measured heat effects to the
enthalpy of the reaction

(NH 2CO)2 NH(s) + 0.7H 2O(l)

= (NH 2CO)2 NH ·0.7H 2O(s) (25)

demonstrates reasonable agreement between the experimental

and the calculated results. Since the thermochemical data were
not taken into account in the estimation of the hydrate
properties, this result confirms the correctness of the proposed
thermodynamic models of phases in the water−biuret system.
Invariant points of the water−biuret phase diagram are given in
Table 4.
The phase diagram of the water−biuret system is presented
in Figure 3. Invariant points of the water−biuret phase diagram
are given in Table 4.
The phase diagram of the urea−water system was studied
repeatedly.1−9 The experimental data are in good agreement, so
they were used with equal statistic weight to obtain model
parameters for the liquid phase. Most authors agree that liquid
in this system is close to ideal. This was disputed in the work of
Shnidman and Sunier,4 but later investigations8 confirmed that
solutions containing more than 60 mol % urea demonstrate
ideal behavior. Our attempts to reproduce the solubility of urea
with Gex = 0 resulted in a satisfactory description of the liquidus
curve over a wide range of compositions but too low eutectic
temperature. Because the eutectic is determined by thermal
analysis with a relatively small error, we used a more complex
two-parametric Margules model for the calculation of phase
equilibria in the urea−water system. The values of AAB and ABA
are listed in the last line of Table 3. The phase diagram is
presented in Figure 4, and the invariant points are given in
Table 4.
3229
Figure 4. Phase diagram of the H2O−(NH2)2CO binary system. Solid
lines indicate the calculated diagram, symbols the experimental data:
○, Speyers et al.;1 ◀, Pinck and Kelly;2 □, Janecke;3 ▽, Miller and
Dittmar;8 △, Chadwell and Politi;5 ◁, Shnidman and Sunier;4 ▷,
Kakinuma;6 *, Babkina and Kuznetsov.9
Ternary Urea−Biuret−Water System. The liquidus
surface of the ternary system was calculated with TernAPI
software.15 As an example, the result of the calculation without
taking into account the binary stoichiometric phases is
presented in Figure 5. In this case one ternary and three
binary eutectics are realized in the urea−biuret−water system.
If both binary compounds I and II are taken into consideration,
three additional peritectics appear. Coordinates of all invariant
points in the ternary system are given in Table 4.
The calculated isothermal sections are shown in Figures 6, 7,
and 8. The sections in Figure 6 illustrate the origin of phase
equilibria in the temperature interval from −5 °C (268.15 K) to
80 °C (353.15 K). Both stoichiometric phases, I and II, were
taken into consideration in the calculation, so these diagrams
show stable phase equilibria in the urea−biuret−water system.
These results can be used for the optimization of the urea and
biuret mutual purification procedure.
dx.doi.org/10.1021/je300832f | J. Chem. Eng. Data 2012, 57, 3225−3232
Journal of Chemical & Engineering Data
Figure 5. Liquidus surface of the ternary urea−biuret−water phase
diagram calculated without taking into account the binary compounds.
The second group of sections shown in Figure 7 illustrates
the variation of phase boundaries at 100 °C (373.15 K) if one
or both binary compounds do not crystallize, for example, due
to kinetic restrictions. These metastable equilibria are very
important for the understanding of crystallization processes in
Figure 6. Calculated isothermal sections of the urea−biuret−water phase diagram: (a) 268.15 K, (b) 298.15 K, (c) 323.15 K, (d) 353.15 K. (I)
(NH2CO)2NH·(NH2)2CO and (II) (NH2CO)2NH·0.7H2O.
3230
Article
the ternary system. It is a well-known fact that metastable
phases may precipitate first during fast cooling. Due to
crystallization of the binary compounds I or II, the liquid
phase field is constricted, so an excess mutual solubility may be
observed under rapid cooling.
To the best of our knowledge, there is only one experimental
study of the ternary phase diagram.26 These results as cited in
ref 12 are superimposed on the calculated phase diagram shown
in Figure 8. The agreement is fair, taking into account that no
details of the experiment are given in ref 12. The composition
of biuret hydrate equimolar in ref 12 should be noted, whereas
the ratio accepted in the present work is (NH2CO)2NH:H2O =
1:0.7.
Additional experimental investigations of the liquid compo-
sition in equilibrium with biuret hydrate and urea would be
useful.
■ CONCLUSIONS
In this work thermodynamic models of the liquid and solid
phases in the urea−biuret−water system were proposed. The
model parameters were optimized based on the available
experimental phase diagrams. It was shown that simple regular
(or subregular) solution models can be used for adequate
description of the phase equilibria in the binary subsystems.
Contrary to assumptions of some authors, the ideal solution
dx.doi.org/10.1021/je300832f | J. Chem. Eng. Data 2012, 57, 3225−3232
Journal of Chemical & Engineering Data Article

Figure 7. Calculated isothermal sections of the urea−biuret−water phase diagram at 373.15 K: (a) without any binary compounds, (b) one
compound, I, in the biuret−urea system, (c) with only biuret hydrate, II, taking into account (d) assuming that both binary compounds I and II are
present.

Figure 8. Isothermal sections of the urea−biuret−water system: (a) 298.15 K, (b) 323.15 K. Dashed lines with circles correspond to the result of ref
12. Compositions are in weight percent.

model is not suitable for the simulation of the eutectic point in
the water−urea system.
The enthalpy and entropy of the urea−biuret binary
compound and of biuret hydrate were estimated using the
solubility data. An experimental confirmation of the proposed
thermodynamic properties is desirable. The estimated enthalpy
of biuret hydrate is in reasonable agreement with the
3231
calorimetric data that were not used for the optimization of
model parameters.
The Gibbs energy of ternary liquid was extrapolated from the
binary subsystems using the Muggiani formalism. The liquidus
surface and isothermal sections of the urea−biuret−water stable
and metastable phase diagram were calculated. According to
our calculations, the solubility of biuret hydrate in water in the
dx.doi.org/10.1021/je300832f | J. Chem. Eng. Data 2012, 57, 3225−3232
Journal of Chemical & Engineering Data Article

presence of urea is smaller than it was proposed earlier; this (21) Kozyro, A. A.; Frenkel, M. L.; Krasulin, A. P.; Simirsky, V.;
result should be confirmed experimentally. Kabo, G. Y. Thermodynamic properties of variuos phase of biuret.

■
Russ. J. Phys. Chem. 1988, 42, 1752−1756.
(22) Muggiani, J.-M.; Gambino, M.; Bros, J.-P. Enthalpies de
AUTHOR INFORMATION formation des alliages liquides bismuth−etain−gallium a 723 K. Choix
Corresponding Author d’une representation analytique des grandeurs d’exces integrales et
partieles de mélange. J. Chim. Phys. 1975, 72, 83−88.
*E-mail: ira@td.chem.msu.ru. (23) Babkina, T. S.; Kuznetsov, A. V.; Miroshnichenko, E. A.;
Funding Paschenko, L. L. Book of Abstracts, XXV-th Int. Chugaev Conference of
The investigations were financially supported by RFBR (Grant Coordination Chemistry, Suzdal, Russia, June 6−11, 2011.
No. 11-03-00499-a and 12-03-31069-mol-a) and the URAL- (24) Baker, L. E.; Pierce, A. C.; Luks, K. D. Gibbs energy analysis of
phase equilbria. Soc. Pet. Eng. AIME J. 1982, 731−742.
CHEM OJSC. (25) Voronin, G. F.; Pentin, I. V. Decomposition of Solid Solutions
Notes of Cadmium, Mercury and Zinc Tellurides. Russ. J. Phys. Chem. A
The authors declare no competing financial interest. 2005, 79, 1572−1578.

■
(26) Babkina, T. S.; Kuznetsov, A. V.; Miroshnichenko, E. A.;
Paschenko, L. L. Thermochemical properties of biuret hydrate. Abs. XXV-
REFERENCES th Int. Chugaev Conference of Coordination Chemistry, Suzdal,
(1) Speyers, C. L. Solubilities of some carbon compounds and Russia, June 6−11, 2011.
densities of their solutions. Am. J. Sci. 1902, 14, 293−302.
(2) Pinck, K. K.; Kelly, M. The Solubility of Urea in Water. J. Am.
Chem. Soc. 1925, 47, 2170−2172.
(3) Janecke, E. Ü ber das System H2O, CO2 and NH3. Z. Elektrochem.
1930, 35, 716−727.
(4) Shnidman, L.; Sunier, A. A. The Solubility of Urea in Water. J.
Phys. Chem. 1932, 36, 1232−1240.
(5) Chadwell, H. M.; Politi, F. W. The Freezing Points of
Concentrated Aqueous Solutions of Urea, Urethan and Acetamide. J.
Am. Chem. Soc. 1938, 60, 1291−1293.
(6) Kakinuma, H. Communication to the Editor: The Solubility of
Urea in Water. J. Phys. Chem. 1941, 45, 1045−1046.
(7) Frèjacques, M. Les bases théoruques de la synthèse industrielle de
l’urée. Chim. Ind. 1948, 60, 22−35.
(8) Miller, F. W. J.; Dittmar, H. R. The Solubility of Urea in Water.
The Heat of Fusion of Urea. J. Am. Chem. Soc. 1934, 56, 848−849.
(9) Babkina, T. S.; Kuznetsov, A. V. Phase Equilibria in Binary
Subsystems of Urea.Biuret. Water System. J. Therm. Anal. Calorim.
2010, 101, 33−40.
(10) Rollet, A.-P.; Cohen-Adad, R.; Hackspill, M. L. Action de la
chaleur sur les mélanges durée et de biuret. C. R. Acad. Sci., Ser. IIC:
Chem. 1949, 18, 199−201.
(11) Rollet, A.-P.; Cohen-Adad, R. Le systeme eau.biuret. C. R. Acad.
Sci., Ser. IIC: Chem. 1951, 2214−2216.
(12) Kucheryavyi, V. I.; Lebedev, V. V. Synthesis and application of
carbamide; Khimiya: Leningrad, 1970; in Russian.
(13) Barone, G.; Cacace, P.; Castronuovo, G.; Elia, V. Excess
enthalpies of aqueous solutions of biuret at 25 °C. Interaction with
urea. Can. J. Chem. 1981, 59, 1257−1260.
(14) Voronin, G. F. Convex functions in the thermodynamics of
heterogeneous substances. Russ. J. Phys. Chem. A 2005, 79, 1890−
1902.
(15) Voskov, A. L.; Voronin, G. F. A universal method for calculation
isobaric−isothermal sections of ternary system phase diagram. Russ. J.
Phys. Chem. A 2010, 84, 525−533.
(16) Glushko, V. P.; Gurvich, L. V., Eds. Thermodynamic properties of
individual substances; Nauka: Moscow, 1978; Vol. 1(2).
(17) Kozyro, A. A.; Dalidovich, S. V.; Krauslin, A. P. Heat capacity,
enthalpy of fusion and thermodynamic properties of urea. Russ. J. Appl.
Chem. 1987, 59, 1456−1459.
(18) Gambino, M.; Bros, J.-P. Capacite calorifique de l’uree et de
quelques melanges eutectiques a base d’uree entre 30 et 140 °C.
Thermochim. Acta 1988, 127, 223−236.
(19) Ruehrwein, R. A.; Huffman, H. M. Thermal Data. XIX. The
Heat Capacity, Entropy and Free Energy of Urea. J. Am. Chem. Soc.
1946, 68, 1759−1761.
(20) Berman, R. G.; Brown, T. H. Heat capacity of minerals in the
system Na2O−K2O−MgO−FeO−Fe2O3−Al2O3−SiO2−TiO2−H2O−
CO2: representation, estimation, and high temperature extrapolation.
Contrib. Mineral. Petrol. 1985, 89, 168−183.

3232 dx.doi.org/10.1021/je300832f | J. Chem. Eng. Data 2012, 57, 3225−3232