American Journal of Analytical Chemistry, 2013, 4, 756-762
Published Online December 2013 (http://www.scirp.org/journal/ajac)
http://dx.doi.org/10.4236/ajac.2013.412092
Open Access AJAC
Terahertz Time-Domain Spectroscopy to Identify and
Evaluate Anomer in Lactose
Satoshi Yamauchi1*, Sakura Hatakeyam1, Yoh Imai2, Masayoshi Tonouchi3
1Department of Biomolecular Functional Engineering, Ibaraki University, Hitachi, Japan
2Department of Electric and Electronic Engineering, Ibaraki University, Hitachi, Japan
3Institute of Laser Engineering, Osaka University, Osaka, Japan
Received November 9, 2013; revised December 2, 2013; accepted December 12, 2013
Copyright © 2013 Satoshi Yamauchi et al. This is an open access article distributed under the Creative Commons Attribution Li-
cense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
ABSTRACT
Lactose powder consisting of α-D-lactose monohydrate and anhydrous β-D-lactose was nondestructively and quantita-
tively evaluated by transmission-type Terahertz time-domain spectroscopy (THz-TDS). An absorption with peak at 39.7
cm−1 (1.19 THz) was assigned to be derived from anhydrous β-D-lactose, in addition to the absorptions due to α-D-
lactose monohydrate with peak at 17.1 cm−1 (0.53 THz) and 45.6 cm−1 (1.37 THz). After deconvolution of the spectra
using Lorentzian, integrated intensities of the absorptions with peak at 39.7 cm−1 and 45.6 cm−1 were uniquely depend-
ent on the weight composition ratio of the α- and β-lactose powder. As a result, the net molar-ratio of the α- and
β-lactose in lactose powder could be precisely evaluated by the integrated intensity ratio. Further, anomer content in
lactose powder extracted from lactose solution was evaluated and the refined and unrefined features were shown by the
evaluation method.
Keywords: THz-TDS; Lactose-Powder; Anomer; α-Lactose; β-Lactose
1. Introduction
Lactose comprising a glucose linked to a galactose abun-
dantly present in milk of most mammals is an important
disaccharide used in foods and pharmaceutical applica-
tions. Two anomers (α-lactose and β-lactose) commonly
exist in the disaccharide powder. Evaluation and control
of the ratio are important to use lactose in foods and
drugs because the α- and β-form show significantly dif-
ferent physicochemical properties, i.e. solubility (β-form
is more soluble than α-form) and hardness (α-form is
harder than β-form), and crystallized shape [1]. Com-
monly, lactose can be prepared as α-lactose monohydrate
and anhydrous β-lactose crystalline in addition to the
amorphous phase of the α- and β-form mixture. The α-
and β-crystals are usually formed in supersaturated lac-
tose solution, where α-lactose can be crystallized with
preventing the β-crystallization at room-temperature but
the β-lactose is condensed above 93.5˚C with extracting
of the α-crystal [2]. Amorphous lactose fabricated by
freeze-drying or spray-drying is crystallized into several
crystal forms such as α-lactose monohydrate [3], anhy-
drous β-lactose [4] and anhydrous crystal with α- and
β-lactose in molar ratios of 5:3 and 4:1 [5]. The crystal-
lization which may enhance both physical and chemical
deterioration [6] is dependent on the composition ratio,
drying process, storage temperature, period and humidity
[7], where the precise evaluation of crystallized α/β-form
ratio is so important to study the crystallization feature.
Commonly, the α-lactose/β-lactose ratio and the crys-
tallization behavior are analyzed by thermogravimetric
analysis (TGA) and differential scanning calorimetry (DSC)
[8,9], Raman spectroscopy [10], Fourier transform infra-
red spectroscopy (FTIR) [11], X-ray diffraction (XRD)
[7,12] etc. However, the TGA and DSC bring to destruc-
tion of sample, and the Raman spectroscopy, FTIR and
XRD can evaluate in significantly thin region limited
within dozens of micrometers. In contrast, THz-spec-
troscopy is expected to be useful for nondestructive char-
acterization of materials with the thickness in millime-
ter-order because THz-electromagnetic wave in very-far-
infrared region is notably absorbed in water but easily
passed through most inorganic and organic materials
comparing to UV-vis-infrared light. Further, it has been
recognized that THz-absorptions based on the intra-mo-
*Corresponding author.

S. YAMAUCHI ET AL. 757
lecular vibrations in organics and/or the inter-molecular
dynamics incorporated hydrogen-bonding are significantly
dependent on molecular- and crystal-structure as demon-
strated for DL-alanine racemic compound comparing to
D- and L-alanine [13], L-phenylalanine comparing to L-
tyrosine [14], and three different retinal isomers with
polyene chain [15]. Many monosaccharides and disac-
charides are also interesting biomolecules for the THz-
spectroscopy since they show the spectral fingerprints
with relatively narrow-bands in THz-region [16,17]. α-D-
lactose also shows the typical absorptions with narrow-
bands in THz-region below 3 THz [18] as well as the
other disaccharides. However, THz-spectroscopy has not
been used to determine the anomer content because the
absorption spectra derived from β-D-lactose has not been
identified.
In this study, transmission-type THz time-domain spec-
troscopy (THz-TDS) is applied to characterize lactose.
Firstly, we focus to identify THz-absorption due to β-D-
lactose, then achieve quantitative evaluation of net molar
ratio of α- and β-lactose in the mixtures using integrated
intensity ratio of the absorptions due to α-D-lactose and
β-D-lactose.
2. Experimental
2.1. THz-TDS System
Transmission-type THz-TDS system as shown in Figure
1 was used to characterize lactose-powers. After a
femto-second laser-light (peak wavelength at 782 nm,
half-width of 87 fs, repetition rate of 48 MHz) was split
into a pump light and a probe light, the pump light
chopped at 1 kHz was focused and irradiated on a THz
emitter consisting of a dipole antenna with 10 μm-gap
space fabricated on LT-GaAs layer and attached on a
hemispherical Si-lens, in which the antenna was biased at
10 V to generate transient current in pico-second order.
The radiated THz-pulse through the Si-lens was focused
Figure 1. Schematic configuration of transmission-type THz
time-domain spectroscopy (THz-TDS) system.
and normally incident to the sample using two off-axis
parabolic metal mirrors. Then the THz-pulse passed
through the sample was introduced to the detector with
the same antenna configuration of the emitter by two
off-axis parabolic mirrors and a hemispherical Si-lens.
When the THz-pulse was introduced to the antenna on
the detector, the probe laser light was simultaneously
irradiated to detect the pulse by sampling technique, where
time-delay of the probe light was controlled by a retro-
reflector and a micro-step stage controller with the step
length of 1 μm. The sampling data were recorded in a PC
after signal-amplification, lock-in noise reduction and
A/D conversion. The recorded pulse data was processed
by a DFT after a Gaussian-window was superposed on
the pulse data to remove aliasing. Humidity and tem-
perature in the THz-TDS system was carefully controlled
below 5% and at 20˚C, respectively, by a dehumidifier
and a heater system to prevent THz-absorptions of water
vapor [19] and thermal fluctuation of samples.
2.2. Preparation of Lactose Sample
Two types of lactose powders purchased from Sigma-
Aldrich Co., Ltd. were used for source materials. The one
was α-D-lactose monohydrate (Lα·H2O: O-β-galactopyra-
nosyl-(1→4)-α-D-glucopyranose monohydrate
(C12H22O11·H2O)) and the other was anhydrous β-D-lac-
tose (Lβ: O-β-galactopyranosyl-(1→4)-β-D-glucopyra-
nose (C12H22O11)), in which a hydrogen- and a hydroxyl-
coordination in the glucose-unit are spatially different in
each other as shown in Figure 2. It should be noted here
that the lactose powders were including the anomer. The
anomer content in the Lα·H2O powder (content of Lβ) and
the Lβ powder (content of Lα·H2O) were about 4% and
Figure 2. Molecular structures of α-D-lactose (α-lactose)
and β-D-lactose (β-lactose).
Open Access AJAC

S. YAMAUCHI ET AL.
758
below 30% in the commercially specifications, respec-
tively. The lactose powers were mixed with various
weight ratios and milled, then filled in a metal aperture
with a hole and compressed to form pellet of 6 mm-di-
ameter and 0.7 mm-thickness with parallel surfaces. The
pellet was placed with the aperture in the THz-TDS sys-
tem and THz-wave was normally incident to the pellet
surface. After the THz-TDS measurement, weight of the
lactose-pellet was measured by an electronic weight-
scale.
2.3. Extraction from Lactose Solutions
Two types of extract processes (refine and unrefined
process) were used to examine the extracting behavior of
Lα·H2O from lactose solution, where the Lα·H2O powder
was dissolved in ultra-pure water with the resistivity
above 18.2 MΩcm. As the refine process, lactose was
crystallized by seeding of the Lα·H2O powder (10 mg)
into a supersaturated solution at 20˚C (1 g-Lα·H2O/3
cc-water) for 24 hrs and dried at 60˚C for 72 hrs in an
incubator after removal the residual solution, then milled
and pressed in the aperture. On the other, lactose was
crystallized from an unsaturated solution (0.3 g-Lα·H2O/3
cc-water) without the seeding, where water in the solu-
tion was gradually vaporized in a dehumidifier for 24 hrs
to form wet powder then milled and pressed in the aper-
ture after dry at 60˚C for 72 hrs in an incubator.
3. Results and Discussions
3.1. Absorption Property of Lactose Powders
Figures 3 show (a) transmission and (b) absorption spec-
tra of Lα·H2O powder (denoted as αP) and Lβ powder (βP),
where the incident THz-wave (Blank) without the lactose
sample is also shown in (a). In these spectra, any absorp-
tion due to water vapor could not be observed in suffi-
ciently low humidity (<5%) and spectrum fluctuation due
to aliasing was successfully removed by Gaussian window
superposed on the THz-pulse. Since THz-wave from 0.2
to 3 THz could be observed in this system but the inten-
sity was exponentially decreased with the frequency, the
frequency useful to analyze the lactose was limited be-
low 1.7 THz because of absorption due to lactose-pellet.
Typical narrow-bands and broad-band increasing with
frequency were observed in the absorption spectra as
shown Figure 3(b). Intense absorptions with peak at 0.53
and 1.37 THz were observed in the Lα·H2O powder (αP),
which were in good agreement to other reports by THz-
TDS [18] and assigned to lactose-active modes origi-
nated from Lα-molecular rotations in Lα·H2O crystal as
shown by first-principles calculations [20]. The absorp-
tions were significantly decreased in the Lβ powder (βP)
but the other absorption with peak at peak 1.19 THz was
clearly observed. Figure 4 shows the absorption spectra
Figure 3. (a) Transmission and (b) absorption spectra of
commercially available α-D-powder (denoted as “αP”) and
β-D-lactose powder (denoted as “βP”). The spectrum de-
noted as “Blank” in (a) was obtained without lactose sam-
ple.
Figure 4. Absorption spectra of the α-D-lactose powder and
the β-D-lactose powder mixtures with various β-D-lactose
powder mixed weight-ratio, where the absorption coeffi-
cient was normalized by the sample weight in mg and the
background broad-band was removed by polynomial func-
tion.
Open Access AJAC

S. YAMAUCHI ET AL. 759
of lactose-pellets consisting of the Lα·H2O powder and Lβ
powder with various weight-content of the Lβ powder
(βP/(αP + βP)) in the pellets, where the broad-band was
removed by polynomial cubic function. It is noted that
the coefficient should be used in the molar coefficient but
was normalized by the weight of pellet because of the
uncertain anomer ratio in the Lα·H2O powder and the Lβ
powder. The intense absorption with peaks at 17.5 cm−1
(0.53 THz) and 45.6 cm−1 (1.37 THz) observed in the
Lα·H2O powder were gradually decreased with the Lβ-
content whereas the absorption with peak at 39.7 cm−1
(1.19 THz) was gradually increased with the rate. It is
not difficult to recognize from the results that the absorp-
tion dominated in Lβ powder is derived from Lβ. Further,
it is expected that the net anomer content in the powders
can be evaluated from the intensity ratio of the absorp-
tions derived from Lα·H2O and Lβ. For the purpose, the
absorption spectra as shown in Figure 4 have to be de-
convoluted to each spectrum. Previously, it was reported
that the lowest-lying absorption with peak at 17.5 cm−1
observed in Lα·H2O powder can be successfully fitted by
Lorentzian [21]. In this work, not only the lowest-lying
absorption but also the other absorptions below 55 cm−1
were successfully fitted by Lorentzian as shown in Fig-
ure 5, where the dot-line depicts a fit by four spectra
with peak at 17.5, 30.0, 39.7 and 45.6 cm−1 (solid-lines)
with the experimentally obtained absorption for a lactose
pellet of 50 wt% Lβ powder (solid-circles). In the spectral
feature, the absorption with peak at 39.7 cm−1 due to Lβ
(FWHM around 7.5 cm−1) was significantly broader than
the two spectra due to Lα·H2O (FWHM below 4 cm−1).
The absorption around 30 cm−1 was also broad with the
FWHM of 7.8 and slightly observed in the pellet with the
Lβ powder weight-ratio above 25%, but the absorption
was not originated from anhydrous Lβ as discussed in
Section 3.3.
Figure 5. Absorption spectrum of the α-D-lactose powder
ted-line, respectively.
th peak at
α
and the β-D-lactose powder mixture with the β-D-lactose
powder mixed weight-ratio of 50%, where the experimen-
tally obtained data, the deconvoluted spectra and a fit by
the spectra are shown by solid-circles, solid-lines and a dot-
3.2. Determination of Anomer Content
The integrated intensities of absorptions wi
−1
45.6 cm (open-circles and denoted as “Iα”) and 39.7
cm−1 (closed-circles and denoted as “Iβ”) were linearly
dependent on the Lβ powder weight-content as shown in
Figure 6, in which the correlative square-factors on the
least squares method were 99.8% and 99.6% for Iα and Iβ
respectively. The results indicate the intensities of Iα and
Iβ for the Lβ powder weight-ratio can be shown by the
below relationships.

IP P1
ααα
xαxα

ww
wαw
w
1P
rP
β β
ββ
αβ α
αxx α
x
αx
 
 


(1)





ww
ww
w
IP1 P
1P 1
rP 1
αα ββ
αβ βα
αβ α
xαxβ
βxx β
x
βx

 

 
(2)
where Pα and Pβ are mixed weight-ratio of the Lα·H2O
powder and the Lβ powder (Pα + Pβ = 1), xα and xβ are net
anomer weight-content in the Lα·H2O powder and the Lβ
powder, αw and βw are absorption coefficient per weight
of Lα·H2O and Lβ, rα and rβ are rate constant of Iα and Iβ
for Lβ weight-ratio, respectively. Then, the rate constant
ratio (r) of rα/rβ for the Lβ-powder weight-content can be
reduced from the Equation (1) and (2) to the follow,
wMM
360 1.05 r
r


wMM
342r



 
  (3)
where αM and βM are molar-absorption co
d
efficient of
Lα·H2O and Lβ, an360 and 342 are molecular weight of
Figure 6. Integrated intensity of absorption derive from
α-D-lactose monohydrate (peak at 45.6 cm−1; openles)
d
circ
and anhydrous β-D-lactose (peak at 39.7 cm−1; solid-circles)
for the β-D-lactose powder mixed weight-ratio in the α- and
β-lactose powders mixture, where the relationships for Iα
and Iβ were obtained by the least-squares method.
Open Access AJAC

S. YAMAUCHI ET AL.
760
Lα·H2O and Lβ. The rate constant ratio of r could be
determined as 0.721 since the r and r were experi
α βmentally
obtained by the results in Figure 6 as 11.07 and 13.36
respectively. On the other, the integrated absorption ratio
of Iα(zα) and Iβ(zβ) in lactose-powder including zα-mol
Lα·H2O and zβ-mole Lβ in the THz-wave pathway should
be described as below,


w
w
Iz

zz
1.05
Iz zr z



 


. (4)
As a result, the net mole-ratio (zβ/zα) in t
be
he sample can
determined by the integrated absorption ratio of Iβ(zβ)/
Iα(zα) and the r-value (0.721). Figure 7 shows net Lβ
molar-content in the pellets with various Lβ weight-con-
tent evaluated by the Equation (4) using the integrated
absorption ratio (Iβ/Iα) and the r-value. In the result, cor-
relative square-factors by the least squares method for
the dependence on the weight-ratio showed an excellent
value as high as 99.8%. The net Lβ molar-content uniquely
increased from 3.9% to 70.9% with the Lβ-powder
weight-ratio indicated anomer ratio was 3.9% in the
Lα·H2O powder and 29.1% in the Lβ powder. The anomer
ratios in the preliminary used powders were in good
agreement to the commercial specifications (Lβ about 4%
in the Lα·H2O powder, Lα·H2O below 30% in the Lα·H2O
powder). The quantitative coincident can be concluded
the analysis using the absorptions with peak at 45.6 cm−1
and 39.7 cm−1 were suitable to determine the anomer
content in lactose powders. It should be mentioned that
the evaluation can be applied to lactose-including sam-
ples with uncertain thickness and density because the
ratio can be determined by the integrated intensity ratio
in the THz-absorption spectrum and the r-value.
Figure 7. Net anhydrous β-D-lactose molar-content for the
β-D-lactose powder weight-ratio in the α- and β-lactose
As ation to determine the α-lactose/β-lactose
powders mixtures, where the net molar-content in the β-
D-lactose powder (βP/(αP + βP) = 1) and the α-D-powder
(βP/(αP + βP) = 0) were evaluated as 71.9% and 3.9% respec-
tively.
3.3. Anomer Content in Extracted Lactose
Powders
demonstra
molar-content ratio, lactose powders extracted from lac-
tose solution were examined by THz-TDS as shown in
Figure 8, in which the absorption spectra after removal
background broad-band and deconvoluted spectra are
shown by dot-lines and solid-lines respectively. It is
noted that the absorption coefficient is shown in cm−1
without normalization by the lactose-weight. Commonly,
it has been recognized that Lα·H2O can be crystallized in
supersaturated lactose solution [22]. Figure 8(a) shows
THz-absorption spectrum of lactose extracted from a
supersaturated solution (1.0 g-Lα·H2O powder/3 cc-water)
by seeing of 10 mg-Lα·H2O powder including Lβ with 3.9
%. The lactose was extracted at 20˚C for 24 hrs and dried
at 60˚C for 72 hrs after removal the solution and then
milled and pressed in the aperture. Absorption spectrum
consisting of intense Lα·H2O-absorption and weak Lβ-
absorption showed a net Lβ-lactose molar-content as low
as 2.8%, which showed the Lα·H2O was refined. In con-
trast, the absorption by lactose powder gradually ex-
tracted from lactose solution without the seeding showed
quite different behavior as shown in Figures 8(b) and (c),
where (a) the water of 3 cc in unsaturated lactose solu-
tion (0.3 g-Lα·H2O powder/3 cc-water) was gradually
vaporized at 25˚C for 24 hrs in a dehumidifier and then
(b) dried at 60˚C for 72 hrs. For the spectrum of Lα·H2O-
absorption with peak at 45.6 cm−1, the FWHM was de-
creased from 5.1 cm−1 to 3.1 cm−1 by the post-anneal. In
contrast, any decrease of the FWHM of Lβ-absorption
with peak at 39.7 cm−1 was not substantially observed
after the post-annealing. The significant difference of the
Figure 8. Absorption spectra of (a) refined lactose iu-
persaturated lactose-solution, (b) as-extracted lactose by
n a s
gradual evaporation of water from unsaturated lactose so-
lution and (c) post-annealed lactose of the sample for (b),
where the dot-lines and the plain-lines show experimentally
obtained spectra and the deconvoluted spectra.
Open Access AJAC

S. YAMAUCHI ET AL. 761
two absorption behaviors by the post-annealing sug-
gested that α-lactose can be crystallized into the hydrates
with excess water although anhydrous crystal is formed
by β-lactose. The peak intensity ratio of Lβ/Lα·H2O was
decreased but the integrated intensity ratio was increased
from 46.2% to 62.7% by the post-annealing because the
absorption of Lα·H2O was narrowed by the annealing. As
a result, the net Lβ-molar ratio was increased from 46.2%
to 62.7% after the annealing.
The evaluated net Lβ-molar ratio of 46.2% in the as-
extracted powder was higher than that in the refined
powder but significantly low comparing to the equilib-
rium ratio of 62.7% in solution [23], which could be
recognized by the solubility of Lα lower than of Lβ [24].
The increased of net Lβ-molar ratio after the post-an-
nealing indicated non-crystallized lactose was crystal-
lized in mostly Lβ during the annealing. It was reported
that collapsed lactose formed from non-collapsed spray-
dried amorphous lactose by exposure in 50% RH for long
time is crystallized in mostly Lβ with some presence of
Lα·H2O at relatively low temperature of 70˚C [8]. Crys-
tallization of the extracted powder in this work seemed to
be similar to the crystallization of collapsed lactose, that
is, collapsed lactose was included in the as-extracted
lactose powder. It should be mentioned that the broad-
band around 30 cm−1 observed in the as-dried lactose was
almost disappeared after the annealing. The decrease of
the absorption and the preference crystallization into Lβ
during the annealing speculates the broad-band is derived
from collapsed lactose.
4. Conclusion
THz-TDS was applied to non-de-
ed by JSPS KAKENHI
esearch Number 24655165.
[1] M. G. Gänzle “Lactose: Crystal-
lization, Hydr Derivatives,” In-
Transmission-type
structive evaluation of anomer content in lactose powder.
Absorption spectra of lactose-pellets consisting of Lα·H2O
and Lβ powders were significantly dependent on the
mixed rate. The absorption with a peak at 39.7 cm−1
(1.19 THz) was assigned to be originated from Lβ in ad-
dition to the fingerprint of Lα·H2O at 17.5 cm−1 (0.53
THz) and 45.6 cm−1 (1.37 THz). Since the integrated
values of the absorption coefficient were uniquely de-
pendent on the powder mixed rate of Lα·H2O/Lβ, the net
anomer content in lactose powder could be precisely de-
termined such as 3.9% and 70.9% for Lα·H2O and Lβ
powders, respectively. Low Lβ-content of 2.8 mol% in
lactose powders evaluated by the THz-TDS analysis
showed the refined feature of Lα·H2O in supersaturated
lactose solution using Lα·H2O-seeds. In contrast, Lβ-
content in lactose powder extracted from unsaturated
lactose solution by gradual water evaporation was in-
creased from 34.4% to 45.3% by post-annealing at 60˚C.
A broad-band with a peak at 30.5 cm−1 was speculated to
collapsed-lactose absorption by the behavior crystallized
in mostly Lβ after the annealing.
5. Acknowledgements
pportThis research was partially su
Grant-in-Aid for Exploratory R
REFERENCES
, G. Haase and P. Jelen,
olysis and Value-Added
ternational Dairy Journal, Vol. 18, No. 7, 2008, pp. 685-
695. http://dx.doi.org/10.1016/j.idairyj.2008.03.003
[2] C. S. Hudson, “Future Studies on the Forms of Milk-
Sugar,” Journal of the American Chemical Society, Vol.
30, No. 11, 1908, pp. 1767-1783.
http://dx.doi.org/10.1021/ja01953a015
[3] C. A. Aguilar and G. R. Ziegler, “
scopic Characterization of Dry Whole M
Physical and Micro-
ilk with Altered
Lactose Content. 2. Effect of Lactose Crystallization,”
Journal of Dairy Science, Vol. 77, No. 5, 1994, pp. 1198-
1204.
http://dx.doi.org/10.3168/jds.S0022-0302(94)77058-2
[4] P. Würsc
tallization of β-Lactose under Elevated Storage Tempe
h, J. Rosset, B. Köllreutter and A. Klein, “Crys-
ra-
se Crystallization in
ture in Spray-Dried Milk Powder,” Milchwissenschaft,
Vol. 39, No. 10, 1984, pp. 579-582.
[5] K. Jouppila, J. Kansikas and Y. H. Roos, “Glass Transi-
tion, Water Plasticization, and Lacto
Skim Milk Powder,” Journal of Dairy Science, Vol. 80,
No. 12, 1997, pp. 3152-3160.
http://dx.doi.org/10.3168/jds.S0022-0302(97)76286-6
[6] E. Berlin, B. A. Anderson and
Vapor Sorption Properties of Various Dried Milks a
M. J. Pallansch, “Water
nd
Whey,” Journal of Dairy Science, Vol. 51, No. 9, 1968,
pp. 1339-1344.
http://dx.doi.org/10.3168/jds.S0022-0302(68)87191-7
[7] N. Drapier-Beche
“Evaluation of Lactose Crystalline Forms by Nondestr
, J. Fanni, M. Parmentier and M. Vilasi,
uc-
tive Analysis,” Journal of Dairy Science, Vol. 80, No. 3,
1997, pp. 457-463.
http://dx.doi.org/10.3168/jds.S0022-0302(97)75957-5
[8] P. Darcy and G. Bu
Drying on the Crystallisation of Amorphous Lactose af
ckton, “The Influence of Heating:
ter
Structural Collapse,” International Journal of Pharma-
ceutics, Vol. 158, No. 2, 1997, pp. 157-164.
http://dx.doi.org/10.1016/S0378-5173(97)00245-7
[9] Á. GombÁs, P. Szabó-Révész, M. Kata, J. G.
I. Erős, “Quantitative Determination of Crystallin
Regdon and
ity of
α-Lactose Monohydrate by DSC,” Journal of Thermal
Analysis and Calorimetry, Vol. 68, No. 2, 2002, pp. 503-
510. http://dx.doi.org/10.1023/A:1016039819247
[10] B. M. Murphya, S. W. Prescottb and I. Larson, “Meas-
urement of Lactose Crystallinity Using Raman Spectros-
copy,” Journal of Pharmaceutical and Biomedical Analy-
sis, Vol. 38, No. 1, 2005, pp. 186-190.
http://dx.doi.org/10.1016/j.jpba.2004.12.013
[11] J. H. Kirk, S. E. Dann and C. G. Blatchf
Definitive Guide to Polymorph Determinatio
ord, “Lactose: A
n,” Interna-
tional Journal of Pharmaceutics, Vol. 334, No. 1-2, 2007,
pp. 103-114.
Open Access AJAC

S. YAMAUCHI ET AL.
Open Access AJAC
762
http://dx.doi.org/10.1016/j.ijpharm.2006.10.026
[12] Md. K. Haque and Y. H. Roos, “Lactose: A Definitive
Guide to Polymorph Determination,” Carbohydr
search, Vol. 340, No. 2, 2005, pp. 293-301.
ate Re-
mination of
[13] M. Yamaguchi, F. Miyamaru, K. Yamamoto, M. Tani and
M. Hangyo, “Terahertz Absorption Spectra of L-, D-, and
DL-Alanine and Their Application to Deter
Enantiometric Composition,” Applied Physics Letters, Vol.
86, 2005, Article ID: 053903.
http://dx.doi.org/10.1063/1.1857080
[14] F. Miyamaru, M. Yamaguchi, M. Tani, M. Hangyo, K.
Yamamoto and K. Tominaga, “THz-Tim
troscopy of Amino Acids in Solid Ph
e-Domain Spec-
ase,” Conference on
Lasers and Electro-Optics, Baltimore, 6 June 2003.
[15] M. Walther, B. Fischer, M. Schall, H. Helm and P. U.
Jepsen, “Far-Infrared Vibrational Spectra of All-Trans,
9-cis and 13-cis Retinal Measured by THz Time-Domain
Spectroscopy,” Chemical Physics Letters, Vol. 332, No.
3-4, 2000, pp. 389-395.
http://dx.doi.org/10.1016/S0009-2614(00)01271-9
[16] P. C. Upadhya, Y. C. Shen, A. G. Davies and E. H. Lin-
field, “Far-Infrared Vibrational Modes of Polycry
Saccharides,” Vibrational Spectroscopy, Vol. 35,
s
No. 1,
talline
2004, pp. 139-143.
http://dx.doi.org/10.1016/j.vibspec.2003.12.010
[17] H. B. Liu and X. C. Zhang, “Dehydration Kinetics of D-
Glucose Monohydrate Studied Using THz Time
Spectroscopy,” Chemical Physics Letters, Vol.
-Domain
429, No.
1-3, 2006, pp. 229-233.
http://dx.doi.org/10.1016/j.cplett.2006.07.100
[18] A. Roggenbuck, H. Schmitz, A. Deninger, I. C. Mayorga,
J. Hemberger, R. Güsten and M. Grüninger
Broadband Continuous-Wave Terahertz Spect
, “Coherent
roscopy on
67-2630/12/4/043017
Solid-State Samples,” New Journal of Physics, Vol. 12,
2010, Article ID: 043017.
http://dx.doi.org/10.1088/13
. De-
(98)00091-0
[19] H. M. Pickett, R. L. Poynter, E. A. Cohen and M. L
lisky, J. C. Pearson and H. S. P. Müller, “Submillimeter
Millimeter, and Microwave Spectral Line Catalog,” Jour-
nal of Quantitative Spectroscopy and Radiative Transfer,
Vol. 60, No. 5, 1998, pp. 883-890.
http://dx.doi.org/10.1016/S0022-4073
hi, R. [20] S. Saito, T. M.Inerbaev, H. Mizuseki, N. Igaras
Note and Y. Kawazoe, “First Principles Calculation of
Terahertz Vibrational Modes of a Disaccharide Monohy-
drate Crystal of Lactose,” Japanese Journal of Applied
Physics, Vol. 45, No. 43, 2006, pp. L1156-L1158.
http://dx.doi.org/10.1143/JJAP.45.L1156
[21] D. G. Allis, A. M. Fedor, T. M. Korter, J. E. Bjarnason
and E. R. Brown, “Assignment of the Lowest-Lying THz
Absorption Signatures in Biotin and Lactose Monohy-
drate by Solid-State Density Functional Theory,” Chemi-
cal Physics Letters, Vol. 440, No. 4-6, 2007, pp. 203-209.
http://dx.doi.org/10.1016/j.cplett.2007.04.032
[22] N. Drapier-Beche, J. Fanni and M. Parmentier, “Physical
0.3168/jds.S0022-0302(99)75510-4
and Chemical Properties of Molecular Compounds of
Lactose,” Journal of Dairy Science, Vol. 82, No. 12, 1999,
pp. 2558-2563.
http://dx.doi.org/1
m.
.org/10.3168/jds.S0022-0302(62)89398-9
[23] T. A. Nickerson, “Lactose Crystallization in Ice Crea
IV. Factors Responsible for Reduced Incidence of Sandi-
ness,” Journal of Dairy Science, Vol. 45, No. 3, 1962, pp.
354-359.
http://dx.doi
[24] P. Walstra, J. T. Wouters and T. J. Geurts, “Dairy Science
and Technology,” 2nd Edition, CRC Press, 2005, p. 754.
http://dx.doi.org/10.1201/9781420028010.ax7