The solubility data of rifamycin S were measured in isopropanol, butyl acetate, and their mixed solvents across the temperature range of 283.15-323.15 K by the gravimetric method. The results demonstrate that the solubility of rifamycin S increases with the increasing temperature in the two pure solvents, and in the mixed solvents, it increases first and then decreases with increasing butyl acetate content. The experimental data of rifamycin S in the mixed solvents were better correlated using the modified Apelblat equation and ideal model equation. Furthermore, the relevant thermodynamic parameters of the dissolution process were determined based on the van't Hoff equation. The obtained dissolution enthalpy and Gibbs free energy are positive in all cases, which indicate that the dissolving process of rifamycin S is endothermic and nonspontaneous. The supersolubility data of rifamycin S were measured by the laser and thermal analytic method. The results demonstrate that the width of the metastable zone of rifamycin S becomes larger with decreasing cooling rate and increasing butyl acetate content. Furthermore, the crystallization process of rifamycin S was optimized on the basis of thermodynamic research. The results showed that when V butyl acetate:V mixed solvent was 0.04, the cooling rate was 0.1 K/min, the stirring rate was 150 rpm, the final crystallization temperature was 283.15 K, and the aging time was 8 h, the purity of rifamycin S crystals could reach 98.5%, and the crystalline yield was 89.6%. After crystallization optimization, the size of rifamycin S crystals increased, and the dissolution in water was improved.
The solubility data of rifamycin S were measured in isopropanol, butyl acetate, and their mixed solvents across the temperature range of 283.15-323.15 K by the gravimetric method. The results demonstrate that the solubility of rifamycin S increases with the increasing temperature in the two pure solvents, and in the mixed solvents, it increases first and then decreases with increasing butyl acetate content. The experimental data of rifamycin S in the mixed solvents were better correlated using the modified Apelblat equation and ideal model equation. Furthermore, the relevant thermodynamic parameters of the dissolution process were determined based on the van't Hoff equation. The obtained dissolution enthalpy and Gibbs free energy are positive in all cases, which indicate that the dissolving process of rifamycin S is endothermic and nonspontaneous. The supersolubility data of rifamycin S were measured by the laser and thermal analytic method. The results demonstrate that the width of the metastable zone of rifamycin S becomes larger with decreasing cooling rate and increasing butyl acetate content. Furthermore, the crystallization process of rifamycin S was optimized on the basis of thermodynamic research. The results showed that when V butyl acetate:V mixed solvent was 0.04, the cooling rate was 0.1 K/min, the stirring rate was 150 rpm, the final crystallization temperature was 283.15 K, and the aging time was 8 h, the purity of rifamycin S crystals could reach 98.5%, and the crystalline yield was 89.6%. After crystallization optimization, the size of rifamycin S crystals increased, and the dissolution in water was improved.
Rifamycin antibiotics
are a class of antibiotics produced by Streptomyces Mediterranean, which have a wide range
of antibacterial effects.[1] They have a
strong influence on Gram-positive bacteria such as Mycobacterium tuberculosis, leprosy-causing bacteria, Streptococcus, and Pneumococcus, especially, drug-resistant Staphylococcus aureus.[2,3] Rifamycin S (C37H45NO12, CAS registry no. 13553-79-2)
is an important precursor for the synthesis of other rifamycin families. Figure shows the molecular
structure of rifamycin S.[4−6]
Figure 1
Molecular structure of rifamycin S.
Molecular structure of rifamycin S.Rifamycin S is generally synthesized from rifamycin
B[7,8] and rifamycin SV. In industrial production, rifamycin
S is often
prepared by acidification of rifamycin S-Na. In 2011, Tang[9] reported a method of synthesizing rifamycin S
from rifamycin SV. First, rifamycin SV fermentation broth was filtered,
and then, the filtrate was oxidized to rifamycin S and then extracted
with butyl acetate. The crude product was washed, concentrated, and
dried to obtain a fine product. In 2016, Zhao Deming[10] reported a process for producing rifamycin S from rifamycin
S-Na. The steps were as follows. First, rifamycin S-Na was reacted
with acid to obtain crude rifamycin S. Second, the acidified soft
water was mixed with the alcohol to wash the crude rifamycin S. Third,
rifamycin S was prepared by recrystallization. In 2017, Li Peng[11] reported a preparation method of rifamycin S
with high purity. The steps were as follows. At first, rifamycin S
was added into the solvent of tetrahydrofuran for dissolution, and
then, the acid alumina into the reactor was added with agitation.
With the agitation process, cyclohexane was slowly added into the
solution until the solid precipitates out. The suspension was agitated
for 5 h until a large amount of solid precipitates out. The filtered
solid was washed with cyclohexane and then dried to obtain bright
red rifamycin S crystals with a purity of 99%.At present, most
of the reports related to rifamycin S are about
its synthesis process,[12] but there are
few studies on the thermodynamics, kinetics, and purification crystallization
process of rifamycin S.[13−15] As an important precursor for
the synthesis of rifampicin and other rifamycin families, the purity,
particle size, and crystal form of rifamycin S directly affect the
production of subsequent drugs. However, in traditional industrial
production, in order to obtain high-purity rifamycin S, it usually
converts the rifamycin S obtained by oxidation of the fermentation
broth rifamycin SV into rifamycin S-Na first and then acidifies rifamycin
S-Na to rifamycin S.[16] The innovation of
this study is that through the crystallization process, high-purity
rifamycin S can be directly obtained, instead of the intermediate
rifamycin S-Na. All the purity data of rifamycin S in the paper were
measured by high-performance liquid chromatography. After the crystallization
process, the purity reached 98.5%, and our research group successfully
synthesized qualified rifampicin with high-purity rifamycin S. In
addition, our research team has developed an effective oxidation separation
method from rifamycin SV to rifamycin S, which improve the purity
of rifamycin S used as a raw material in subsequent crystallization.[17] In the process of rifamycin S preparation, butyl
acetate is used as the extractant and isopropanol is used as the crystallization
solvent.[10] Therefore, in the purification
crystallization system of this study, isopropanol–butyl acetate
was selected as the solvent, and the solubility of rifamycin S was
measured in nine different ratios of isopropanol–butyl acetate
mixed solvents.Therefore, this work first studied the solubility
of rifamycin
S in isopropanol, butyl acetate, and their mixed solvents across the
temperature range of 283.15–323.15 K by the gravimetric method.[18] The experimental data were correlated with the
modified Apelblat equation and ideal model equation. According to
the van’t Hoff equation, relevant thermodynamic data of dissolution
were obtained. According to the obtained thermodynamic data, the crystallization
process of rifamycin S was optimized by cooling crystallization. The
effects of final crystallization temperature, cooling rate, stirring
rate, aging time, seed crystal, and butyl acetate content on the crystallization
of rifamycin S were investigated. The results showed that the crystalline
form of rifamycin S was improved, the particle size and distribution
of the crystals were optimized, and the purity of the crystals was
improved, which provided strong support for the subsequent synthesis
of rifampicin equality drugs.
Results and Discussion
Thermodynamic Properties of Rifamycin S
Solubility
of Rifamycin S
The solubility
data of rifamycin S in butyl acetate, isopropanol, and their mixed
solvents were determined in the temperature range of 283.15–323.15
K. The mole fraction of rifamycin S in organic solvents can be calculated
using the following eq (19)wherein x1 is
the mole fraction of rifamycin S; m1, m2 is the mass of rifamycin S and the solvent; M1 and M2 is the
molar mass of rifamycin S and solvent, respectively.Figure shows the solubility
curve of rifamycin S in mixed solvents with different butyl acetate
volume fractions.
Figure 2
Molar fraction solubility of rifamycin S in different
solvent compositiona
(Vbutyl acetate:Vmixed solvents).
Molar fraction solubility of rifamycin S in different
solvent compositiona
(Vbutyl acetate:Vmixed solvents).The temperature range was selected according to the operable range
of the production process. In this paper, five points were selected
for measurement with a temperature difference of 10 °C. Each
measurement point was carried out for more than three sets of experiments,
and it was found that it was enough to get a good pattern. The solubility
of rifamycin S in isopropanol and butyl acetate solvents increases
with the increasing temperature. At the same temperature, the solubility
of rifamycin S in butyl acetate is much greater than that in isopropanol.
In a isopropanol–butyl acetate mixed solvent, the solubility
of rifamycin S increases first and then decreases with increasing
butyl acetate content. When the proportion of butyl acetate is 0.8–0.9,
the solubility is maximum. After the volume fraction of butyl acetate
is greater than 0.8, the solubility of rifamycin S in the isopropanol–butyl
acetate mixed solvent is greater than that in either of the pure solvents,
indicating that the mixed solvent has a solubilizing effect in this
case. Considering the dissolution performance alone, choosing the
right solvent ratio and proper dissolution temperature can not only
reduce heating energy consumption but also save solvents.
Rifamycin S Solubility Correlation
The modified Apelblat
equation is a semiempirical model, which has
widely been used to correlate the mole fraction solubility against
temperature.[20,21]The ideal model equation
is a commonly
used equation describing the solid–liquid equilibrium state
based on the principle of thermodynamics.[22]here A, B, C, a, and b are
parameters of this equation. T is the absolute temperature. x is the mole fraction solubility of the solute.The
relative deviation (RD) between the calculated value of the
solubility model and the experimental data can be calculated by eqThe root-mean-square deviation (RMSD) can be obtained from eq wherein N is the number of
experimental data points; x is the experimental solubility
value; xcal is the calculated solubility
value.The obtained RD is shown in Table , and the relevant parameters and RMSD are
shown in Table . It
can be seen from Table that the R2 obtained by the Apelblat
equation was greater
than 0.994, and the RMSD values were less than 10–4 under different solvent ratios. From the overall results, the fitting
effects of the Apelblat equation were better than the ideal model
equation.
Table 1
Experimental and Calculated Solubility
Data of Rifamycin S in Different Solvents
Apelblat
ideal model
Vbutyl acetate:Vmixed solvent
T/K
x
xcal
100 × RD
xcal
100 × RD
0
283.15
1.33 × 10–4
1.22 × 10–4
8.2707
1.29 × 10–4
3.0075
293.15
2.35 × 10–4
2.46 × 10–4
–4.6809
2.55 × 10–4
–8.5106
303.15
5.01 × 10–4
5.02 × 10–4
–0.1996
4.84 × 10–4
3.3932
313.15
1.03 × 10–3
1.03 × 10–3
0.0000
8.80 × 10–4
14.5631
323.15
2.12 × 10–3
2.12 × 10–3
0.0472
1.54 × 10–3
27.3585
0.1
283.15
2.85 × 10–4
2.99 × 10–4
–4.9123
2.58 × 10–4
9.4737
293.15
5.01 × 10–4
5.11 × 10–4
–1.9960
5.40 × 10–4
–7.7844
303.15
1.00 × 10–3
9.57 × 10–4
4.3000
1.08 × 10–3
–8.0000
313.15
1.91 × 10–3
1.94 × 10–3
–1.5707
2.06 × 10–3
–7.8534
323.15
4.21 × 10–3
4.21 × 10–3
0.0950
3.77 × 10–3
10.4513
0.3
283.15
1.01 × 10–3
9.70 × 10–4
3.9604
9.65 × 10–4
4.4554
293.15
1.84 × 10–3
1.88 × 10–3
–2.1196
1.89 × 10–3
–2.7174
303.15
3.46 × 10–3
3.53 × 10–3
–1.9653
3.55 × 10–3
–2.6012
313.15
6.53 × 10–3
6.44 × 10–3
1.4242
6.40 × 10–3
1.9908
323.15
1.14 × 10–2
1.14 × 10–2
–0.3509
1.11 × 10–2
2.6316
0.5
283.15
4.47 × 10–3
4.47 × 10–3
0.0447
4.15 × 10–3
7.1588
293.15
5.91 × 10–3
6.03 × 10–3
–1.9459
6.30 × 10–3
–6.5990
303.15
8.85 × 10–3
8.61 × 10–3
1.3860
9.30 × 10–3
–5.0847
313.15
1.28 × 10–2
1.29 × 10–2
–1.0234
1.34 × 10–2
–4.6875
323.15
2.03 × 10–2
2.03 × 10–2
0.1232
1.89 × 10–2
6.8966
0.7
283.15
1.02 × 10–2
9.96 × 10–3
1.0693
9.92 × 10–3
2.7451
293.15
1.27 × 10–2
1.28 × 10–2
–0.4409
1.29 × 10–2
–1.5748
303.15
1.64 × 10–2
1.63 × 10–2
0.2500
1.65 × 10–2
–0.6098
313.15
2.07 × 10–2
2.06 × 10–2
0.3188
2.08 × 10–2
–0.4831
323.15
2.60 × 10–2
2.61 × 10–2
–0.2769
2.58 × 10–2
0.7692
0.8
283.15
2.30 × 10–2
2.31 × 10–2
–0.3348
2.24 × 10–2
2.6087
293.15
2.34 × 10–2
2.32 × 10–2
0.2833
2.38 × 10–2
–1.7094
303.15
2.42 × 10–2
2.42 × 10–2
–0.0950
2.51 × 10–2
–3.7190
313.15
2.62 × 10–2
2.60 × 10–2
0.6298
2.65 × 10–2
–1.1450
323.15
2.88 × 10–2
2.87 × 10–2
0.2188
2.78 × 10–2
3.4722
0.85
283.15
2.42 × 10–2
2.42 × 10–2
–0.1901
2.34 × 10–2
3.3058
293.15
2.46 × 10–2
2.44 × 10–2
0.4919
2.49 × 10–2
–1.2195
303.15
2.50 × 10–2
2.54 × 10–2
–0.0906
2.64 × 10–2
–5.6000
313.15
2.73 × 10–2
2.74 × 10–2
–0.2491
2.78 × 10–2
–1.8315
323.15
3.04 × 10–2
3.03 × 10–2
0.3618
2.92 × 10–2
3.9474
0.9
283.15
2.46 × 10–2
2.47 × 10–2
–0.3821
2.43 × 10–2
1.2195
293.15
2.48 × 10–2
2.47 × 10–2
0.2621
2.49 × 10–2
–0.4032
303.15
2.52 × 10–2
2.52 × 10–2
0.0278
2.55 × 10–2
–1.1905
313.15
2.59 × 10–2
2.60 × 10–2
–0.5174
2.61 × 10–2
–0.7722
323.15
2.72 × 10–2
2.72 × 10–2
–0.1728
2.67 × 10–2
1.8382
1
283.15
2.02 × 10–2
2.02 × 10–2
–0.0990
1.99 × 10–2
1.4851
293.15
2.05 × 10–2
2.05 × 10–2
0.1951
2.05 × 10–2
0.0000
303.15
2.09 × 10–2
2.09 × 10–2
0.0000
2.10 × 10–2
–0.4785
313.15
2.15 × 10–2
2.16 × 10–2
–0.2326
2.16 × 10–2
–0.4651
323.15
2.24 × 10–2
2.24 × 10–2
0.0893
2.21 × 10–2
1.3393
Table 2
p Arameter Regression
Results of the Apelblat Equation and Ideal State Equation Model
modified
Apelblat equation
ideal
model equation
Vbutyl acetate:Vmixed solvent
A
B
C
R2
104 × RMSD
a
b
R2
104 × RMSD
0
–40.74
–4420.74
8.35
0.9990
0.0699
11.10
–5679.31
0.9899
0.2682
0.1
–66.13
–3348.62
12.29
0.9946
0.2474
13.41
–6136.83
0.9916
0.2120
0.3
–83.69
–1266.17
14.39
0.9995
0.5999
12.82
–5593.27
0.9988
0.1546
0.5
–418.72
15674.68
63.40
0.9986
0.9556
6.74
–3461.23
0.9863
0.7452
0.7
–67.82
995.26
10.57
0.9970
0.7214
3.12
–2189.74
0.9955
0.1888
0.8
–212.91
8994.43
31.42
0.9974
0.9166
–2.04
–498.13
0.9793
0.6957
0.85
–221.50
9377.28
32.71
0.9969
0.8247
–1.97
–505.11
0.9742
0.9359
0.9
–106.29
4432.56
15.40
0.9952
0.8158
–2.94
–220.45
0.9540
0.3098
1
–67.59
2672.51
9.61
0.9968
0.3130
–3.10
–230.71
0.9424
0.2000
Analysis of Dissolution
of Rifamycin S
The Van’t Hoff equation[23] can
be used to calculate the standard molar enthalpy of solution (ΔsolnHm°). It can be known from eq that the ΔsolnHm° of the solute can be obtained by calculating the slope
of the graph of ln x ∼ 1/T. The standard molar dissolution Gibbs free energy (ΔsolnGm°) and standard molar dissolution entropy (ΔsolnSm°) can be obtained from eqs and 9.[24] In this study, because the temperature range
was small (283.15–323.15 K), the value of the heat of dissolution
can be regarded as a constant.[25] These
equations are as follows.where Tm refers
to the average of all test temperatures. R is the
gas constant, for which the value is 8.314 J/(K·mol).The
calculated thermodynamic data of ΔsolnHm°, ΔsolnGm°, and ΔsolnSm° of
rifamycin S are shown in Table .
Table 3
Relative Solution Thermodynamic Properties
of Rifamycin S in Different Solvent Composition (VButyl Acetate:VMixed Solvent)
Vbutyl acetate:Vmixed solvent
ΔsolnHm°
ΔsolnGm°
ΔsolnSm°
0
77.0103
18.9944
191.3768
0.1
73.8132
17.2296
186.6523
0.3
67.2754
14.1976
175.0876
0.5
41.6314
11.7965
98.4162
0.7
26.3380
10.3525
52.7312
0.8
6.0740
9.2876
–10.6008
0.85
6.0020
9.1622
–10.4245
0.9
6.0294
9.2867
–10.7448
1
2.7750
9.7224
–22.9175
It can be seen from Figure and Table that the values of ΔsolnHm° and
ΔsolnGm° of
rifamycin S in isopropanol–butyl acetate mixed solvent are
positive in all cases, indicating that the dissolution process of
rifamycin S is endothermic and nonspontaneous in this experimental
study.
Figure 3
Linear correlation of ln x ∼ (1/T – 1/Tm).
Linear correlation of ln x ∼ (1/T – 1/Tm).
Metastable
Region of Rifamycin s in Isopropanol
and Butyl Acetate
With isopropanol as the solvent, the metastable
zones of rifamycin S at different cooling rates were measured at a
stirring rate of 150 rpm, as shown in Figure .
Figure 4
Mesostable zone of rifamycin S in isopropanol
at different cooling
rates.
Mesostable zone of rifamycin S in isopropanol
at different cooling
rates.When rifamycin S has a lower cooling
rate in isopropanol, the width
of the metastable region is larger, that is, it has a larger operating
range, which is more conducive to the growth of the crystal. When
the cooling rate is higher, the width of the metastable region is
smaller, which is more conducive to crystal nucleation. Therefore,
in order to cultivate large and uniform crystals, the crystallization
should take place at a lower cooling rate to maintain a low energy
transfer rate and the solution should be kept in the metastable zone
during the crystallization process and avoid spontaneous nucleation.The width of the metastable zone of rifamycin S in solvent with Vbutyl acetate:Vmixed solvents at 0, 10, and 30% was, respectively, measured at a stirring rate
of 150 rpm and a cooling rate of 0.1 K/min, which is shown in Figure
Figure 5
Width of the mesostable
zone of rifamycin S by temperature difference
in different solvent compositions (Vbutyl acetate:Vmixed solvent).
Width of the mesostable
zone of rifamycin S by temperature difference
in different solvent compositions (Vbutyl acetate:Vmixed solvent).The width of the metastable region becomes larger while the
proportion
of butyl acetate in the mixed solvent increases, which is more beneficial
to the crystal growth. As the extraction agent in industrial production,
a little butyl acetate is left in the crystallization solvent. The
investigation of solubility of rifamycin S in different mixed solvents
can provide some theoretical reference for the extraction, purification,
and production of rifamycin S.It can be seen from Figures and 5 that the width of the metastable
zone is smaller at lower temperatures and larger at higher temperatures.
This is because the solubility of rifamycin S does not change significantly
with temperature, so the width of the metastable zone becomes larger
at higher temperatures.
Crystallization
Process of Rifamycin S
Effect of Final Crystallization
Temperature
on the Crystallization of Rifamycin S
From the solubility
curve of rifamycin S in isopropanol, it can be seen that its solubility
decreases with the decreasing temperature. Therefore, it can be seen
from Figure when
the final crystallization temperature gradually decreased, the solubility
of rifamycin S decreased, and the supersaturation (i.e., the crystallization
driving force) was constantly provided during the crystallization
process. The small particles formed by stirring and crushing became
the nucleus of secondary nucleation, thereby crystallizing to form
small crystal grains, causing the average particle size to decrease.
Figure 6
Particle
size distribution at different crystallization final temperatures.
Particle
size distribution at different crystallization final temperatures.Purity and yield of rifamycin S at different crystallization
final
temperatures.Particle size distribution of rifamycin S at
different cooling
rates.Crystal forms of rifamycin S at different cooling
rates.Purity and yield of rifamycin S at different
cooling rates.Particle size distribution of rifamycin
S at different stirring
rates.Crystal forms of rifamycin S at different
stirring rates.Purity and yield of
rifamycin S at different stirring rates.Particle
size distribution of rifamycin S under different aging
times.Purity and yield of rifamycin S under
different aging times.Particle size distribution
of rifamycin S under different Vbutyl acetate:Vmixed solvents.Particle size distribution of rifamycin S at different seed additions.Figure shows that
the purity of rifamycin S gradually decreased with the decreasing
final crystallization temperature, and it was considered that the
solubilities of impurities decreased with the decreasing temperature.
The yield of rifamycin S was an increasing trend. When the final crystallization
temperature was below 283.15 K, the yield did not change much because
the solubility of rifamycin S in the solvent decreased with the decreasing
crystallization temperature. Based on the results, considering the
energy and time savings, the final crystal temperature was selected
to be 283.15 K.
Figure 7
Purity and yield of rifamycin S at different crystallization
final
temperatures.
Effect of the Cooling
Rate on Crystallization
of Rifamycin S
As can be seen from Figure , when the cooling rate was lower, the metastable
zone width was larger, and the driving force needed for crystal growth
was provided slowly. Saturation took place slowly, which kept the
solution in the metastable zone during crystallization, avoiding spontaneous
nucleation. Therefore, Figure shows that at a lower cooling rate, larger and uniform crystals
can be obtained.
Figure 8
Particle size distribution of rifamycin S at
different cooling
rates.
As can be seen from Figure ,with the decreasing cooling rate, the particle
size of rifamycin S gradually increased, and the shape changed from
long rods to flakes, the aspect ratio decreased, and the thickness
increased. This was because when the cooling rate was large, the driving
force needed for crystal growth was provided fast and the nucleation
rate was much higher than the growth rate. When the cooling rate was
slow, the crystals were fully grown, so the crystalline nature was
better.
Figure 9
Crystal forms of rifamycin S at different cooling
rates.
Considering the crystal size distribution, crystal shape,
and purity
(Figure ), the selected
cooling rate was 0.1 K/min.
Figure 10
Purity and yield of rifamycin S at different
cooling rates.
Effect
of the Stirring Rate on Rifamycin
S Crystallization
Figure shows that without stirring, rifamycin S had a wide
particle size range with two higher peaks. When the stirring rate
was 100 or 150 rpm, the particle size range of rifamycin S was narrow.
This showed that providing appropriate agitation strength was beneficial
to obtain a rifamycin S product with uniform particle size.
Figure 11
Particle size distribution of rifamycin
S at different stirring
rates.
From Figure , without
stirring, each side of rifamycin S was irregular and the particle
size difference was large. When the stirring rate was 225 rpm, the
crystals were broken too much. When the stirring speed was 150 rpm,
the crystal was complete and the particle size was uniform. This was
because when the stirring rate was or was not too low, the solvent
and the solute could not be fully mixed and allowed to make contact,
and local supersaturation levels would be formed, resulting in uneven
crystal size and two peaks. When the stirring rate was too high, the
crystals would be broken. In addition, the crystal had an obvious
curved surface without stirring, but it did not exist when there was
stirring, indicating that stirring had changed the crystalline nature.
Figure 12
Crystal forms of rifamycin S at different
stirring rates.
It can be seen from Figure that when other conditions were constant, the yield
of rifamycin S increased with the increasing stirring rate, but it
increased slowly when the stirring speed exceeds 100 rpm. It was considered
that when the stirring rate was low, the metastable zone of rifamycin
S was wide, which led to the incomplete crystallization of rifamycin
S. Considering comprehensively, the stirring rate was selected as
150 rpm.
Figure 13
Purity and yield of
rifamycin S at different stirring rates.
Effect of Aging Time
on Crystallization
of Rifamycin S
It can be seen from Figure that when other conditions were constant,
the particle size of rifamycin S increased with increasing aging time,
indicating that after reaching the final crystallization temperature,
the crystallization in the solution made small particles grow into
large grains. Proper aging time would facilitate to form larger rifamycin
S crystals.
Figure 14
Particle
size distribution of rifamycin S under different aging
times.
Figure shows that the yield and purity of rifamycin S were basically
unchanged with the increase of aging time, indicating that after a
period of aging, the system had reached crystallization equilibrium.
Considering comprehensively, the aging time of 8 h was more appropriate.
Figure 15
Purity and yield of rifamycin S under
different aging times.
Effects of Different Solvent Composition
(VButyl Acetate:VMixed Solvent) on Crystallization of Rifamycin S
Figure shows
thatwith the decreasing proportion of the butyl acetate in the mixed
solvent, the volume particle diameter of rifamycin S gradually decreased,
the particle size was more uniform. It can be seen from Figure that when the content of butyl
acetate was relatively low in the mixed solvent, with the butyl acetate
content increasing, the metastable zone width became larger. Therefore,
rifamycin S was not completely crystallized, resulting in a smaller
particle size.
Figure 16
Particle size distribution
of rifamycin S under different Vbutyl acetate:Vmixed solvents.
As can be seen from Table ,with the increasing proportion of butyl
acetate in the mixed solvent, the purity of rifamycin S increased
slightly, but the yield decreased. From the solubility of rifamycin
S obtained in the isopropanol–butyl acetate mixed solvent obtained
from the foregoing, it can be known that when the proportion of butyl
acetate in the mixed solvent was small, the solubility of rifamycin
S increased rapidly with the increasing butyl acetate content, so
the yield of rifamycin S decreased. Considering comprehensively, it
was better when Vbutyl acetate:Vmixed solvent was 4%.
Table 4
Purity and Yield of Rifamycin S under
Different Solvent Composition (VButyl Acetate:VMixed Solvents)
experiment
number
Vbutyl acetate:Vmixed solvents/%
purity/%
yield/%
1
0
98.2
88.7
2
4
98.4
87.4
3
10
98.3
80.9
Effect of Seeds on Crystallization of Rifamycin
S
Figure shows that when other conditions were the same, the addition of
rifamycin S seeds could increase the particle size of the rifamycin
S crystal. With the amount of rifamycin S seeds increasing, the particle
size of the rifamycin S crystal did not increase much. When adding
seed crystals in the metastable zone, the seeds were equivalent to
the nucleus of rifamycin S, which changed the width of the metastable
zone and made the crystallization orderly, which was beneficial to
the crystallization of large particle size. Because the total amount
of rifamycin S in the solution was constant, the particle size would
not increase continuously with the increasing amount of seed.
Figure 17
Particle size distribution of rifamycin S at different seed additions.
Table shows that
when the other conditions were the same, with the increasing amount
of rifamycin S seeds, the purity of rifamycin S did not change much,
and the yield increased, but the change of yield was small when the
amount of seeds reached 0.5%.
Table 5
Purity and Yield
of Rifamycin S at
Different Seed Additions
experiment
number
seed amount/%solvents/%
purity/%
yield/%
1
0
98.4
87.4
2
0.5
98.5
89.5
3
1
98.5
89.6
4
2
98.4
89.4
Optimization
Results of Rifamycin S
The optimized experimental method
is as follows. First, 3.72 g of
crude rifamycin S was added into butyl acetate solvent. Then, the
solution was transferred to a rotary evaporator to evaporate 42.3
mL of butyl acetate. Second, isopropanol was added to 191 mL and transferred
to the double-jacketed glass vessel and then crystallized to 283.15
K (adding 1% seeds at 315.15 K) at a cooling rate of 0.1 K/min and
a stirring rate of 150 rpm. Finally, the crystal system was kept static
for 8, and then, the crystalline product was analyzed and detected.The results of the experiment are as follows. The purity of the
rifamycin S crystal after optimization was 98.5%, and the yield was
89.6%. The particle size distribution diagram is shown in Figure . The SEM images
are shown in Figure , in which 19a is crude rifamycin S, and 19b,c are the optimized products under different
magnifications, respectively. The dissolution rate in water is shown
in Figure .
Figure 18
Particle
size distribution of rifamycin S after optimization.
Figure 19
SEM image of rifamycin S.
Figure 20
Dissolution
profile of rifamycin S in water.
Particle
size distribution of rifamycin S after optimization.SEM image of rifamycin S.Dissolution
profile of rifamycin S in water.It can be seen that after optimization, the crystalline nature
of rifamycin S undergoes a great change. The crystalline form of rifamycin
S was very broken before crystallization. After crystallization, the
particle size of rifamycin S crystals increased significantly, and
the aspect ratio decreased. The crystal presented a clear block structure,
and the particle size was more uniform. The purity of the optimized
rifamycin S could reach 98.5%. At the same dissolution time, the dissolution
rate in water was higher than before, which indicates that the optimized
rifamycin S has higher solubility in water, and the optimized rifamycin
S was more conducive to subsequent experiments.
Conclusions
In this work, the solubility of rifamycin S
in pure isopropanol,
pure butyl acetate, and their mixed solvents was measured by using
the gravimetric method from 283.15 to 323.15 K. The solubility data
of rifamycin S increases with the increasing temperature in the two
pure solvents. At the same temperature, the solubility of rifamycin
S in pure butyl acetate is much greater than that in pure isopropanol.
The solubility of rifamycin S increases first and then decreases with
the increasing butyl acetate content. When the volume fraction of
butyl acetate reaches 0.8–0.9, the solubility is maximum. When
the volume fraction of butyl acetate is greater than 0.80, the solubility
of rifamycin S in the mixed solvent is greater than the solubility
in either of the pure solvents, indicating that the mixed solvent
has a solubilizing effect in this case. The experimental solubility
data were correlated by the modified Apelblat equation and ideal model
equation, and the relevant thermodynamic parameters of the dissolution
process were determined based on the van’t Hoff equation. It
was found that the dissolution process of rifamycin S in isopropanol–butyl
acetate mixed solvents is endothermic and nonspontaneous. Furthermore,
the supersolubility data of rifamycin S at different cooling rates
and solvent compositions were measured by the laser and thermal analytic
method. The width of the metastable zone of rifamycin S becomes larger
with decreasing cooling rate and increasing butyl acetate content.Based on the thermodynamic research, the crystallization process
of rifamycin S was optimized. The results showed that when Vbutyl acetate:Vmixed solvent was 0.04, the cooling rate was 0.1 K/min, the stirring rate was
150 rpm, the final crystallization temperature was 283.15 K, and the
aging time was 8 h; the purity of rifamycin S crystals could reach
98.5%, and the crystallization yield was 89.6%. After crystallization
optimization, the particle size of the rifamycin S crystal increased
and became more uniform, and its dissolution in water also increased,
which improved its solubility in water. This research will facilitate
subsequent experimental operations and provide theoretical guidance
for the overall optimization of the rifampicin synthesis process.
Experimental Section
Materials
Rifamycin
S was provided
by a pharmaceutical manufacturer and further purified in the laboratory.
The organic solvents (isopropanol, butyl acetate) used in the experiment
were analytical-grade reagents with the mass fraction purity higher
than 99.5%, determined by gas chromatography. All the organic solvents
were used in experiments without further purification. All reagents
were supplied by Sinopharm Chemical Reagent Co., Ltd.The low-temperature
thermostatic water bath was supplied by Shanghai Shunyu Hengping Scientific
Instrument. The stirrer was supplied by Shanghai Shenshun Biological
Technology Co., Ltd. The vacuum pump was supplied by Shanghai Dayan
Instrument Co., Ltd. The rotary evaporator was supplied by Provided
by Shanghai Yarong Biochemical Instrument Factory. The spectrophotometer
(SPECORD 210 PLUS) was supplied by Jena Analytical Instrument Company
of Germany. The fField emission scanning (NOVA NanoSEM450) electron
microscope was supplied by American FEI company.Figure is a
schematic for the solubility and supersolubility measurement apparatus.
Figure 21
Schematic
for the solubility and supersolubility measurement apparatus.
1-Laser transmitter; 2-condenser tube 3-solvent dropping tube; 4-thermometer;
5-crystallizer; 6-laser receiver; 7-magnetic stirrer; 8-laser intensity
display; 9-constant temperature device.
Schematic
for the solubility and supersolubility measurement apparatus.
1-Laser transmitter; 2-condenser tube 3-solvent dropping tube; 4-thermometer;
5-crystallizer; 6-laser receiver; 7-magnetic stirrer; 8-laser intensity
display; 9-constant temperature device.
Determination of Thermodynamic Properties
In this work, the gravimetric method was used to measure the solubility
of rifamycin S. The experimental steps are as follows. At first, excess
amounts of rifamycin S were added into the solvent contained in a
double-jacketed glass vessel. The temperature of the glass vessel
was controlled by a thermostat with an accuracy of ±0.05 K. The
suspension was agitated for 8 h to realize the solid–liquid
equilibrium. Then, the agitation was stopped. After that, the upper
clear saturated solutions were withdrawn and filtered by a sand core
funnel and dried in a vacuum oven at 323.15 K until the total weight
was constant. At the same time, the undissolved materials were filtered
by a weighed sand core funnel and dried in a vacuum oven at 323.15
K until the total weight was constant. An electric balance with an
accuracy of ±0.0001 g was used to measure the masses of the samples.The supersolubility of rifamycin S was studied by laser and indirect
thermal analysis. The organic solvents included isopropanol, butyl
acetate, and their mixed solvents. First, quantitative amounts of
the solvent and rifamycin S were added into the crystallizer. At the
same time, the laser recorder was debugged, and the material in the
crystallizer was heated to a temperature 5 K higher than the solubility
temperature of rifamycin S and maintained for a period of time to
ensure that the solute was completely dissolved. Second, the temperature
was cooled at a fixed cooling rate, and the supersolubility curve
was determined according to the change of the light intensity. To
ensure the accuracy of the experimental data, three parallel experiments
were conducted in each group, and the average value was taken.
Optimization of the Crystallization Process
A single
variable method was used to investigate the effects of
different final crystallization temperature, cooling rate, stirring
rate, aging time, solvent system component content, and seed addition
amount on the rifamycin S crystals. At first, 3.72 g of rifamycin
S and 150 g of the solvent were added into the crystallizer. At the
same time, the thermostat and the agitator were opened. When the temperature
had risen to 326.15 K, stirring was carried out at a constant temperature
for 30 min. Then, the solution was cooled at a certain cooling rate
and stirring rate to different final crystallization temperatures.
Then, the agitation was stopped, and the system was kept static for
different aging times at a constant temperature. Finally, the sample
obtained after suction filtration was dried in a vacuum oven for 48
h.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9
Figure 10
Figure 11
Figure 12
Figure 13
Figure 14
Figure 15
Figure 16
Figure 17
Figure 18
Figure 19
Figure 20
Figure 21