close

Вход

Забыли?

вход по аккаунту

?

Monitoring In Situ Growth and Dissolution of Molecular Crystals Towards Determination of the Growth Units.

код для вставкиСкачать
COMMUNICATIONS
monomers is probably not the main degradation process.[']
Other degradation products are instead obtained through OH
radical attack on the heterocyclic base moiety.[''] These experiments point out a sensitivity that is lower than that of classical,
chemical reference methods (Fricke) .[' 'I Unfortunately, doses
that are sufficient for the direct reading of our display are much
higher than those that produce a biological response in living
organisms.
However, the sensitivity of the method could be improved by,
for example, addition of sensitizers,['21use of different oligonucleotides, and experiments with liquid crystals formed in the
presence of additives such as polyamines or basic polypeptides,
which do not inhibit the formation of the cholesteric phase and
might better simulate a cellular situation." 31 An alkaline medium also favors cleavage of the sugar-phosphate backbone once
the base has been damaged.['21 Considering that the pitch before
irradiation is about 58 pm, the cell can be miniaturized to approximately 300 pm, a length that is sufficient for reading five
fingerprint lines.
To our knowledge, the effect of ionizing radiation on liquid
crystals of DNA and guanosine analogs had not yet been studied, and there is no information available on the dependence of
the cholesteric pitch on radiation. All published data in this field
deal with the change of thermodynamic and dielectric properties
after irradiation of cholesterics made of cholesterol derivative~.['~]
Experimental Section
Calf thymus DNA (Sigma) was sonicated as described in ref. [4]. The DNA samples
(32 wt % in water) were prepared by sealing a drop of solution between two cover
slides with the commercially available SureSeal system (thickness 0.2 mm) as a
spacer. For each dose two samples were exposed, and the blank pitch was determined with 12 samples. The dimer d(GpG).Na was synthesized and purified following a previously described procedure [7]. Samples of the d(GpG) cholesteric phase
(4.8 wt% in water) were prepared, sealed in flat glass microslides (Vitro Dynamics,
thickness 0.3 mm), and exposed to y radiation (60Co source, dose rate:
38 Gymin-') for different times corresponding to doses between 1 and 20kGy
(three slides were used for each irradiation time, and the experiment was doubled;
the blank pitch was determined over 18 different slides). The microslides were kept
in a magnetic field (O.R T) for 4 h to form and align the fingerprint lines [3b]. A Zeiss
microscope (Standard 16) was used for determining the pitch.
Received: December 12, 1996 [Z9882IE]
German version: Angew. Chem. 1997,109,989-991
Keywords: biosensors * deoxyguanosine
y radiation * liquid crystals
- DNA structures
*
[I] C. Robinson, Tetrahedron 1961, 13, 219-234.
[2] G. Gottarelli, G. P. Spada, A. Garbesi in Comprehensive Suprumolecular
Chemistry, f i r / . 9 (Eds.: J. L. Atwood, J. E. D. Davies, D. D. MacNicol, F.
Vogtle. J.-M. Lehn. J.-P. Sauvage, M. W. Hosseini), 1st ed., Elsevier, Oxford,
1996. pp. 483-506.
[ 3 ] a) F. Livolant. J. Phys. 1986, 47, 160-1616; b) R. Brdndes, D. R. Kearns,
Biorhemi.w)>1986, 25, 5890-5895.
[4] G. Gottarelli, G. P. Spdda, P. Mariani, M. M. de Morais, Chirality 1991, 3,
227-232.
[5] M. M. Girdud-Guille. Int. Rev. Cyfol. 1996, 166, 59-101.
[6] a) L. P. Canderds, S. Steenken, J. A m . Chem. SOC.1992,114,699-704; b) T.
Melvin, S. W. Botchway, A. W. Parker, P. O'Neill, ihid. 1996, f18, 1003110036.
[7] S. Bonazzi, M. Capobianco. M. M. de Morais, A. Garbesi, G. Gottarelli, P.
Am. Chem. SOC.1991,
Mariani, M. G. Ponzi Bossi, G. P. Spada, L. Tondelli, .l
113, 5809- 5816.
[8] When S'(pG).Na (0.9 wt%) was added to a solution of d(GpG).Na (6.85 wt%
in water), the pitch of the cholesteric mesophase decreased from 4 5 k 6 pm to
17+1 pm.
[9] S. Steenken, Chem. Rev. 1989,89, 503-520.
[lo] a) S. Rdoul, M. Berger, G. W. Bnchko, P. C. Josh, B. Morin, M. Weinfeld, J.
Cadet, J. Chrm. Soc. Perkin Truns. 2 1995, 371-381, and references therein;
b) M. Dizdaroglu, Biochemistry 1985, 24, 4476-4481.
Angew. Chem. Int. Ed. EngI. 1997, 36, No. 9
0 VCH
[ l t ] A. Dutreix, A. Bridier in The Dosimefry of Ionizing Radiation, Vol. 1 (Eds.:
K. R. Kase, B. E. Bjarngard, F. H. Attix), Academic Press, Orlando, 1985,
pp. 163-169.
[12] D. A. Dunn, V. H. Lin, I. E. Kochevar, Biochemistry 1992,3/, 11620-11 625.
[13] J. Pelta, D. Durand, J. Doucet, F. Livolant, Biophys. J. 1996, 71, 48-63.
[14] a) S. J. Klosowicz, 2. B. Alfassi, Mol. Cryst. Liq. Cryst. 1994,239,181-193; b)
S. L. Srivastava, R. Dhar, Rudiat. Phys. Chem. 1996, 47, 287-293.
Monitoring In Situ Growth and Dissolution of
Molecular Crystals: Towards Determination of
the Growth Units**
David Gidalevitz, Robert Feidenhans'l,* Sophie Matlis,
Detlef-M. Smilgies, Morten J. Christensen, and
Leslie Leiserowitz*
Molecular crystal growth in solution takes place at the crystal-fluid interface. The local structure at this boundary plays a
primary role in determining the composition, growth, and habit
of the crystal. Until recently experimental in situ studies of crystal growth made use mainly of optical methods, which provided
information not only on the micron level['] but also on
monomolecular steps.['. 31 With the advent of atomic force
microscopy (AFM)14' it became possible to monitor the crystalliquid interface on the subnanometer scale.[' -']
Over the last decade grazing incidence X-ray diffraction
(GID) has proven to be a powerful tool for obtaining information on surfaces at the atomic and molecular levels.r99 The
technique was developed for studies of metal and semiconductor
surfaces under ultrahigh vacuum['*] but, due to the penetrating
power of X-rays, can also be used for interfaces such as Langmuir layers on water[". 'I or electrolyte
14] In this
study we applied GID to probe the surfaces of organic, molecular crystals and their growth and dissolution interfaces.
As model systems we chose crystals of p-alanine
(+H,NCH,CH,CO;)
and a-glycine (+H,NCH,CO;). Both
are composed of zwitterionic molecules that are interlinked
through strong N-H . . .O hydrogen bonds to form centrosymmetric bilayers (Figure 1). We thus anticipated that the surface
layers would be sufficiently smooth and well-ordered to detect
a GID signal. We addressed principal questions of crystal
[*I
"1
[''I
[**I
Prof. R. Feidenhans'l, Dr. D.-M. Smilgies,"' Dr. M. J. Christensen
Department of Solid State Physics
Riso National Laboratory
DK-4000 Roskilde (Denmark)
Fax : Int. code + (42)37-0115
e-mail: robert@risoe.dk
Prof. L. Leiserowitz, Dr. D. GidaIevitz,''''
Dr. S. M a t h
Department of Materials and Interfaces
The Weizmann Institute of Science
76100 Rehovot (Israel)
Fax: Int. code +(8)934-4138
e-mail: csles@weizmann.weiann.ac.il
New address: Experiments Division, European Synchrotron Radiation Facility, Grenoble Cedex (France)
New address: Physics Dept, University of Pennsylvania (USA)
We are grateful to Ada Yonath and her group of the Max-Planck-Insitut fur
strukturelle Molekularbiologie, Hamburg for use of laboratory facilities and
to the staff of HASYLAB for assistance. We thank Meir Lahav, Isabelle
Weissbuch, and Ivan Kuzmenko for discussions. We acknowledge financial
support from the Minerva Foundation, the Danish National Research Council, and the Fund for Basic Research of the Israel Academy of Sciences and
Humanities.
Verlugsgesellschu~mbH. 0-69451 Weinheim. 1997
0570-0833~97/3609-09S5
$17.50+ .SOjO
955
COMMUNICATIONS
growth, namely, the molecular arrangement of the
growing crystal surface
and the structure of the
building units that dock
onto them from solution.
It is possible to glean information on the number of
molecules making up the
building units by comparison of the theoretical
growth under vacuum and
the experimental morphol0gy.[I5However, for
crystal grown from solution it is generally difficult
to assess the contribution
I
C
of solvent in determining
Figure 1. Packing arrangement of p-alathe morphology. On comnine (top) and a-glycine crystals (bottom)
viewed along the c and a’ axes, respectively.
parison of the “theoretiThe adjacent layers 1 and 2 as well as
cal” morphology of a3 and 4 are related to each other by a center
glycine
and that obtained
of inversion and are bound though strong
by sublimation, Berkohydrogen bonds; layers 1 and 3 as well as 2
and 4 are related by a twofold screw axis
vitch-Yellin proposed that
that is parallel to h.
the glycine molecules form
cvclic dimers in the gas
phase.[”] The absence of large {Old) faces for a-glycine obtained from aqueous solution was interpreted in terms of docking of cyclic dimers.[“I Myerson et a1.[21-23]
employed Gouy
interferometry measurements of diffusion coefficients and evaluation of concentration gradients in a column of supersaturated
glycine solution to obtain an average size of the glycine cluster
of about 1.8 molecules. These results are in agreement with the
characterization of the a-glycine growth interface from AFM
and phase-measurement interferometric microscopy by Carter
et al.[241They obtained a lower limit of 1 nm for the step size on
the (010) glycine face during growth from a saturated aqueous
solution, which is close to the thickness of the hydrogen-bonded
bilayer. However, the assignment of the (010) glycine surface
structure is ambiguous when only AFM measurements are considered, since the possible surface layer structures 1 and 2, or 3
and 4 (Figure 1, bottom) cannot be distinguished.
We show here that GID and AFM may be combined to advantage to determine the structure with near atomic resolution
of the crystal-fluid interface of p-alanine and a-glycine during
growth and dissolution. This method provides information on
the crystal growth units.
P-Alanine, which crystallizes in the orthorhombic space
group Pbca ( a = 9.9, b =13.8, c = 6.1 A, Z = 8),[25,261was
grown from an aqueous solution to about 8 x lOx6mm’; it
exposes two dominant {OIO} faces and eight { I l l } side faces.
The crystal is composed of hydrogen-bonded bilayers that are
juxtaposed along the b axis by van der Waals C-H . . .H-C
contacts (Figure 1, top). The crystal can be cleaved parallel to
the plane of the bilayer. In principle, the cleaved face may contain the equivalent layers 1 or 3 (in which CH, groups are
exposed), layers 2 or 4 (in which CO; and NH; groups are
exposed), or a combination of these two.
The surface of a freshly cleaved (010) face of 8-alanine was
extremely smooth. Indeed, AFM
on such a
surface yielded molecular resolution with unit-cell dimensions
a = 9.2L-0.9, c = 6.1 k0.5 8, (Figure 2a), which are close to the
bulk values. In situ dissolution of a freshly cleaved (010) surface
of b-alanine in 1-hexanol (Figure ~ C ) [ ’ ~and
I 1-pentanol was
monitored by AFM at room temperature.
956
0 VCH Yerlagsgesellschaft mhH, 0-69451
Weinheim. 1997
Fignre2. AFI images of the {OlO}
. . face c p-alanine. a) Molecular resolution on
the freshly cleaved surface (top) and the results of a Fourier transformation along
the lines shown (bottom). The lattice periods of 9.18 8, and 6.07 A are very similar
to the a and c unit-cell axes of 8-alanine obtained from X-ray diffraction measurements. h) Height profile of a f?-alaninecrystal surface during dissolution in 1-hexanol. c) Snapshots following dissolution of a freshly cleaved {OlO} face in 1-hexanol. The time elapsed after addition to the solution is given in the bottom right
corners.
a-Glycine crystallizes in the monoclinic space group P2,/n
(a=5.1, b=11.8, c = 5 . 5 & ~ = 1 1 2 ” ,Z=4).[291 Crystals
grown from a pure, aqueous solution exhibit bipyramidal morphologyexpressing {IIO), {011}, and (010) faces. As in thecase
of p-alanine, a-glycine can be easily cleaved parallel to the hydrogen-bonded bilayer to yield {OlO} faces with two possible
surface terminations. A freshly cleaved (010) face of glycine was
molecularly smooth (Figure 3b). Dissolution of a freshly
cleaved (010) face of glycine in I-propanol (Figure3c) and
growth in a saturated solution in ethanol/water (3/1, Figure 4),
was monitored by AFM at room temperature.
A series of snapshots of the same region of the surface was
taken during these AFM measurements. Both dissolution and
growth proceeded as step flow with an average minimal step
height of 7kO.5 8, for /I-alanine (Figure 2b) and 5.7k0.5 8, for
a-glycine (Figure 3c). We may conclude from the p-alanine and
a-glycine crystal structures that this corresponds to two molecular layers.
A cell was constructed to measure synchrotron X-rays scattering from the crystal-solution interface (Figure 5). The measurements were performed at the BW2 and W1 wiggler beamlines at the Hamburg Synchrotron Radiation Laboratory
(HASYLAB, Germany) .r301 The sharp termination of a crystal
at its surface gives rise to rods of diffuse scattering in the direction of the surface normal; they pass through the bulk Bragg
0570-0833197/3609-0956 $17.50+ S O / O
Angew. Chem. Inr. Ed. Engl. 1997, 36, No. 9
COMMUNICATIONS
Figure 3. AFM images of the (030) face of cr-glycine. a) Freshly cleaved surface on
a micron scale. b) Molecular resolution on the freshly cleaved surface (top) and the
results of a Fourier transformation along the lines shown (bottom). The lattice
periods of 5.07 A and 5.45 8, are very similar to the a and c unit-cell dimensions ol
r-glycine obtained from X-ray diffraction measurements. c) Snapshots following
dissolution of the freshly cleaved (010) face of r-glycine in 1-propauol. The height
profile for the top, left picture is indicated.
Figure 4. AFM snapshots of the same region on the (010) face of a-glycine during
its growth from a saturated solution of ethanol/water (3/1), and the molecular
resolution (g) of one such surface (h). Snapshot (h) lies between (d) and (e).
reflections. These rods are called crystal truncation rods
(CTR) ,[3 and their shape between the Bragg points contains
information about the molecular structure at the surface, such
as changes in interlayers distances, atomic rearrangements, and
surface roughness.["] In particular, these rods are sensitive to
the changes in surface termination during growth.
Cleaved (010) crystal surfaces of p-alanine and a-glycine were
prepared in the same way as for the AFM measurements. Each
Angew. Chem. Int. Ed. Engl. 1997. 36,.2Vo.9
specimen crystal was glued
to the top surface of a copper block in the cell and
aligned so that the exposed
{Ole} surface was parallel
to the block surface. For
crystal growth or dissolution experiments a drop of
solution was spread onto
the cleaved surface. A
stretched, polypropylene
film was then tightly
pressed onto the solution
to keep the liquid layer as
thin as possible and to prevent its evaporation. Pure
ethanol was used for dissoFigure 5. Schematic view of the cell containing the specimen crystal, which is
lution of the p-alanine
placed on a copper block in contact with
(010) face, and a saturated
a Peltier thermoelectric device to control
solution in ethanollwater
crystal temperature. The crystal surface is
(3/3) for growth of the
exposed t o a thin, fluid layer that is then
covered with a polypropylene film
glycine (010) surface.
(61m). N, gas is pumped through the
X-ray structure factor
cell to prevent water precipitation at low
calculation^^^^
yielding
temperatures.
CTRs from molecularly
smooth (010) glycine and
p-alanine faces in which the surface atoms were placed at the
terminated bulk positions reveal dramatic differences depending on surface termination (Figure 1).Therefore, GID measurements are a sensitive tool for probing surface composition during in situ crystal growth or dissolution.
Measurements of the (Ok4) CTR were made on a dry, freshly
cleaved (010) face of j-alanine at 0 "C (Figure 6a). The data
points (open squares) lie close to the calculated curve (solid line)
for a crystal terminated at layers 1 and 3 (Figure 1, top). The
calculated CTR curve (dashed, red line) for a crystal terminated
at layers 2 and 4 is very different. This proves that the cleavage
occurs between the hydrogen-bonded bilayers. Measurements
for all samples were performed in high-symmetry directions
along the surface to search for higher order reconstructions of
the surface, but none were found.
Dissolution of the same (010) face of j-alanine was monitored by repeating the GID measurements after spreading a
drop of pure ethanol on the surface. From the results of our
AFM measurements of glycine and p-alanine dissolution in different solvents, we estimated that the crystal lost approximately
1000 molecular layers during the experiment. The measured
CTR profile (Figure 6a; filled, red rhombs) again corresponded
to a 0-alanine surface with exposed layers 1 or 3.[321The width
of rocking curves along the CTR did not change as a function
of time, which means that the crystal preserved its quality.
The (Okl) CTR of a pure, cleaved (010) face of a-glycine was
measured at room temperature (Figure 6b). The CTR profile
(open squares) coincided with the model curve (solid line) for a
crystal terminated at layers 1 or 3 (Figure 1, bottom). Again,
the calculated curve (dashed, red line) for a crystal terminated
at layers 2 or 4 is very different. A drop of a saturated solution
in ethanol/water was spread on the surface, and the diffraction measurements were repeated (Figure 6b). From the information on growth rate derived from independent AFM measurements. we estimate that the {OlO} face grew by approximately 100-200 molecular layers during the two hours of X-ray
data collection. Once again the observed data (filled, red
rhombs) are very similar to that measured on the dry, cleaved
crystal (open squares). A minor decrease in the width of the
[C VCH Verlriggesellschuft rnhH 0-69451 Weinheim,1997
0570-0833j97/3609-0957$17.50+ .50/0
957
COMMUNICATIONS
The results described here are compatible with a model in
which molecules of 8-alanine and glycine effectively dock or
leave the crystal surface as cyclic, hydrogen-bonded dimers. It is
possible that this is preferred over docking as monomers, since
the former is less solvated. This growth model is in agreement
with previous arguments and experiments and also suggests that
precipitation of the cr- and y-polymorphs of glycine may depend,
to some extent, on the behavior of glycine in solution. Thus,
crystallization of the a-form from an aqueous solution would be
favored by the presence of cyclic dimers, whereas the
which does not contain cyclic dimers, crystallizes from acid or
base solutions that disrupt cyclic-dimer formation of glycine.
Received: July 22, 1996 [Z9361E]
German version: Angew. Chem. 1997,109,991 -995
Keywords: amino acids * atomic force microscopy
growth * surface analysis
KFigure 6. a) The (Ok4)CTRs from a 8-alanine (010) surface during dissolution in
pure ethanol. b) The (Okl)CTRs from an n-glycine {OlO) surface during growth
from a saturated solution of ethanol/water (3/1).Open squares: data points measured on the pure dry (010)cleaved surface; filled, red rhombs: data points measured in situ during the dissolution of 8-alanine and growth of a-glycine; solid line:
calculated model for the crystal terminated at layers 1 and 3 (Figure 1); dashed, red
line: calculated model for the crystal terminated at layers 2 and 4 (see Figure 1).
k is the (continuous) Miller index along the CTR;fls the structure factor.
rocking scans of the CTR was observed after growth, indicating
that the surface quality was marginally improved during
growth. This is consistent with AFM observations (Figure 4)
that clearly showed improved smoothness of the glycine crystal
surface during growth.
GID experiments on dissolution of the (010) face of p-alanine
and growth of the (010) face of cr-glycine, which showed that the
surface termination of the crystals did not change during X-ray
data collection, are consistent with the AFM measurements
(Figures 2-4). Moreover, the termination was the same as after
cleavage of the crystals. In other words, within the time frame
of the dissolution and growth experiments, the crystal {OIO}
surfaces did not show surface truncations different from the
layers 1 or 3, in which CH, groups are exposed. Otherwise the
observed data points would not all lie on a line Consistent with
the model CTR curve for layers 1 and 3, but somewhere between
this curve and that for layers 2 and 4.
958
0 VCH Erlagsgesellschaft mbH, 0-69451 Weinheim, 1997
crystal
[l]I. Sunagawa, Current Topics in Materials Science, Vol. 10 (Ed.: E. Kaldis),
1982,North-Holland, Amsterdam, p. 353; P. GSrnert, F. Voigt, Current Topics
in Materiab Science, Vol. ti (Ed.: E. Kaldis), 1984,North-Holland, Amsterdam, p. 1 ; A. A. Chernov. Contemp. Phys. 1989,30, 251.
121 K. Tsukamoto. J. Cryst. Growth 1983,61, 199; K. Onuma, K. Tsukamoto,
1. Sunagawa. ibid. 1990,100,125.
[31 L. A. M.J. Jetten, B. van der Hoek, W. J. P. van Enckevort, J. Cryst. Growth
1983,62,603.
[41 G. Binnig, C.F. Quate, C. Gerber, Phys. Rev. Lett. 1986,56,930.
[5]P. E. Hillner, S. Manne, A. J. Gratz, P. K. Hansma, Ultramicroscopy 1990,42,
1387.
[61 S.D.Durbin, W. E. Carlson, J. Cryst. Growth 1992,122,71.
171 P.E. Hillner, A. J. Gratz, S. Manne, P. K.Hansma, Geology 1992,20,359.
[81 P. E.Hillner, S. Manne, P. K. Hansma, A. J. Gratz, Farad. Discuss. 1993,191.
[9] R.Feidenhans’l, Surf: Sci. Rep. 1989,10,105.
[lo]I. K. Robinson, D. J. Tweet, Rep. Prog. Phys. 1992,55, 599.
[11]D. Jacquemain, S. Grayer Wolf, F. Leveiller, M. Deutsch, K. Kjaer, J. AlsNielsen, M. Lahav, L. Leiserowitz. Angew. Chem. 1992, 106. 134; Angew.
Chem. Int. Ed. Engl. 1992,31, 130.
[121 J. Als-Nielsen, D. Jacquemain, K. Kjaer, F. Leveiller, M. Lahav, L. Leiserowitz, Phys. Rep. 1994,246,251.
[I 31 Synchrotron Techniques in lnterfacial Electrochemistry (NATO AS1 Ser,
C 1994,432).
[14]B. M.Ocko, J. Wang, A. Davenport, H. Isaaks, Phys. Rev. Lett. 1990,65,1466.
[IS]P. Hartman, Crystal Growth, North-Holland, Amsterdam, 1973.
[16]L. J. W. Shimon, M. Vaida. L. Addadi, M. Lahav, L. Lelserowitz in Organic
Crystal Chemistry (Eds.: J. B. Garbarczyk, D. W. Jones), University Press,
Oxford, 1991, p. 74.
[17]J. L. Wang, L. Leiserowitz, M. Lahav, J. Phys. Chem. 1992,96,15.
[18]R. F. P. Grimbergen, P. Bennema in Crystal Growth of Orgariic Materials
(Eds.: A. S. Myerson, D. A. Green, P. Meenan), ACS Conf. Proc. Ser. 1996,
p. 28.
[19]Z.Berkovitch-Yellin, J. Am. Chem. Soc. 1985,107,8239.
[201 I. Weissbuch, R. Popovitz-Biro, M. Lahav, L. Leiserowitz, Acra Crystallogr. B
1995,51, 115.
[21]Y C. Chang, A. C. Myerson, A I C h E l 1986,32,1746.
1221 A. S . Myerson, P. Y Lo, J Cryst. Growth 1990,99,1048.
[23]R. M. Ginde, A. S. Myerson, J. Cryst. Growth 1992,116,41.
[24]P. W Carter, A. C. Hillier, M. D. Ward, J. Am. Chem. SOC.1994,If6,944.
[25] P. Jose, L. M. Pant, Acta Crystallogr. 1964,18, 806.
[261 E. Papavinasam, S. Natarajan, N. S. Shivaprakash, Int. J Peptide Protein Res.
1986,28,525.
(271 Conventional AFM imaging was performed with a Nanoscope I11 instrument
(Digital Instruments). The sample was scanned in the contact mode with a
commercial Si,N, tip attached to a cantilever with a spring constant of
0.38N m- I . Images were acquired at room temperature (in air and in solution)
in a fluid cell with a capacity of 0.1 cm3 (zero flow rate). Large-scale images
were taken under minimum force as allowed by feed-back sensitivity (= 10 nN
in air and % 1 nN in solution).
[28] AFM measurements were performed during dissolution of the (010) faces of
8-alanine (1 -pentanol and l-hexanol) and a-glycine (ethanol. 1-propanol, 1-butanol, and l-pentanol). Since the results were very similar for these solvents we
present here dissolution of p-alanine in l-hexanol and growth of a-glycine in
l-propanol.
[29] J.-P. Legros, A. Kvick, Arta Cr).stallogr. B 1980,36,3052.
were mounted on the six-circle diffractometew which were de[301 The
diffraction, at the beamlines w1 and Bw2. The
signed for surface
surface was aligned with a laser so that the surface normal was parallel to the
$ 17.50+ .50/0
OS70-0833/97/3609-0958
Angews. Chem. Int. Ed. Engl. 1997,36,No. 9
(u axis, and a constant angle of incidence was kept thoroughout the measure-
ments. X-ray wavelengths of 1.319 8, for fi-glycine and 1.24 8, for fi-alanine
were chosen. At every point along the Bragg rod an w scan rocking around the
surface normal was performed to obtain integrated intensities, which were
corrected for Lorentzian and polarization factors and variations in the active
area to give structure factor intensities [9]
[31] I. K. Robinson. Phys. Rev. B 1986, 33, 3830.
[32] Measurement of one CTR lasted about 2-3 hours. Each data point corresponds to the integrated intensity of a rocking curve, which took approximately 2 min.
[33] Y. Iitaka, Acta CrwaNogr. 1961, 14, 1
Surface X-ray Scattering Study of Stereospecific
Adsorption of Additives onto the Surface of a
Molecular Crystal Grown from Solution**
David Gidalevitz, Robert Feidenhans’l, and
Leslie Leiserowitz*
Nucleation, growth, and habit of solution-grown molecular
crystals are strongly affected by the nature of the
and the presence of molecular additive^.[^-^] This is primarily a
result of the interaction between the “foreign” molecules and
the crystal faces. The effect of such interactions on the structure
of the crystal was deduced from knowledge of the crystal structure and macroscopic properties such as morphology and symmetry of the resulting crystal,[61but it has not been possible to
observe the surface-bound “foreign” molecules directly.[’]
Recently various experimental methods have been introduced
for examining the surfaces of three-dimensional molecular crystals at the subnanometer scale and thereby probing the structures of foreign molecules adsorbed on such surfaces. For example, atomic force microscopy was used to characterize the
structure of an organic diacid on a layered clay surface.[g1X-ray
diffraction is a very powerful technique for studying surfaces
and does
and interfaces on the atomic and molecular levelr”.
not require handling of the samples under vacuum. So far the
technique has mainly been applied to surfaces of metals and
semiconductors, but it has recently also been used for characterizing mineral surfaces.[”] One restriction when applying the
diffraction techniques for the study of adsorbed foreign molecules is that the surface has to be very smooth, and the adsorbing molecules have to be ordered with respect to the host
crystal.
a-Glycine, which crystallizes in the monoclinic space group
P2,/r1,“~1
was chosen as a model substrate system. The structure
is composed of bilayers parallel to the ac plane (Figure 1, top)
in which the molecules are interlinked through strong hydrogen
[‘I
[ ‘1
[‘*I
Prof. L. Leiserowitz, Dr. D. Gidalevitz“’
Department of Materials and Interfaces
The Weizmann Institute of Science
76100 Rehovot (Israel)
F a x : Int code +(8)934-4138
e-mail : csles(aweizmann.weizmann.ac.il
Prof. Robert Feidenhans’l
Department of Solid State Physics, Risvl National Laboratory, Roskilde
(Denmark)
New address: Physics Department, University of Pennsylvania, Philadelphia
(USA)
We thank Meir Lahav, Isabelle Weissbuch, and Ivan Kuzmenko for discussions. We acknowledge financial support from the Minerva Foundation, the
Danish National Research Council, and the Fund for Basic Research of the
Israel Academy of Sciences and Humanities.
Angew. Chem. h i . E d Engl. 1997, 36, No. 9
Figure 1. Top: packing arrangement of the a-glycine crystals viewed along a*. The
symmetry-related layers 1-4 as well as the (010) and (010) faces are indicated. The
N-H - . - 0hydrogen bonds within and between layers are shown as dashed lines.
Layers 1 and 2 as well as 3 and 4 are interlinked by three-center hydrogen bonds
(N-H...O,O) . The top bilayer shows one type of hydrogen bond, the bottom
bilayer the other. Bottom: proposed packing arrangement of the (010) glycine
surface regrown in the presence of (Qmethionine (9 wt%). The ordered (S)-methionine molecules are bound through a pseudo center of inversion to glycine. The
conformers A and B each form separate domains on the (070) glycine surface.
bonds. Bipyramidal crystals are obtained from an aqueous solution of pure glycine. Experiments have ~ h o w n [ that
~ - ~enan~
tiomerically pure a-amino acid additives are enantioselectively
adsorbed onto the (010) faces[’41of glycine and affect their
growth and morphology. Additives of (R)-a-amino acid inhibit
growth along the +b direction, and the resulting pyramidal
crystal has a large (010) face; (S)-a-amino acids inhibit growth
along the -b direction to yield enantiomorphous pyramids with
(OTO) bases. This observation was explained as follows: An (R)
additive can substitute a glycine molecule only at sites 1 or 3 of
the (010) face (Figure 1, top) causing growth inhibition along
b, and, by symmetry, ( S ) additives can occupy sites 2 and 4 on
the (OTO) face inhibiting growth along - b.
To facilitate detection of layer of adsorbed a-amino acid additive by surface X-ray diffraction and to distinguish between
adsorption of a chiraI-resolved additive on the opposite (010)
and (070) faces, (S)-methionine (CH,SCH,CH,CH(NH:)C0;)was used as an additive. It should inhibit growth only at
the (010) face, and the sulfur atoms enhance the contrast in the
X-ray images.
A crystal of a-glycine was grown to approximately
8 x 10 x 6 mm3, dried, and cleaved along the central ac plane to
yield two halves with large (010) and (070) faces. The two halves,
each about 3 mm thick, were placed in a vessel with their cleaved
faces exposed to the same saturated, aqueous, solution of
glycine with (S)-methionine (9 wt%) as additive. The two crystals were grown further for approximately four hours, removed,
and dried with filter paper. They now revealed distinct differences. The half with the large (OTO) face maintained its original
thickness, whereas the other had grown by about 3 mm at the
exposed (010) face.
+
0 VCH Verlagsgesellschafr mbH, 0-69451
Weinheim, 1997
0570-083319713609-09598 17 SO+ ,5010
959
Документ
Категория
Без категории
Просмотров
4
Размер файла
2 312 Кб
Теги
unit, crystals, monitoring, molecular, towards, growth, determination, dissolution, situ
1/--страниц
Пожаловаться на содержимое документа