当前位置：文档库 › Far-infrared-spectra-of-amino-acids_2005_Chemical-Physics

Far-infrared-spectra-of-amino-acids_2005_Chemical-Physics

Far-infrared spectra of amino acids

Adriana Matei,Natalia Drichko

*,1,

Bruno Gompf,Martin Dressel

1.Physikalisches Institut,Universita ¨t Stuttgart,Pfa?enwaldring 57,D-70550Stuttgart,Germany

Received 25October 2004;accepted 13April 2005

Available online 21July 2005

Abstract

The far-infrared absorption of polycrystalline samples of 18amino acids has been measured in the spectral range from 10to 650cm à1.The assignment of the observed bands is based on comparing the spectra of di?erent amino acids and normal modes calculations present in literature.The absorption peaks can be grouped in ?ve regions that are characteristic for all the specimens.In the region 480–650cm à1COO àrocking/wagging/bending modes are observed.NH t3torsion is dominant in the range 380–480cm à1.Between 270and 380cm à1the CC a

N deformation mode dominates,followed by a region between 220and 270cm à1speci?c for COO àvibrations.Below 220cm à1a one-to-one correspondence of absorption frequencies to certain modes (COO àtorsion,NC a C deformation,NH t3torsion,and omnipresent H bonds)is no more possible.ó2005Elsevier B.V.All rights reserved.

Keywords:Amino acids;Far-infrared absorption

1.Introduction

Proteins are basic constituents of all living organisms and composed of organic molecules,called amino acids which are joined covalently by peptide bonds.The DNA contains the genetic information that dictates the spe-ci?c sequence of amino acids.All the 20a -amino acids found in peptides and proteins,except for proline,con-sist of carboxylic acid (–COOH)and amino (–NH 2)functional groups attached to the same tetrahedral car-bon atom (a C).Organic side chains (R)linked to the same a C atom distinguish one amino acid from another.By means of a polymerization reaction (condensation),amino acids can form chains of di?erent length (in general more than 100)that represent the proteins backbone.

Due to their important biological function,amino acids have frequently been a subject of spectroscopic studies.Nevertheless,the information about them re-mains incomplete:unlike mid-infrared studies (see for instance [1]),there are still gaps in the far-infrared (FIR)characterization of amino acids.In 1935,Heintz performed the ?rst representative investigation of the far-infrared absorption of some 2amino acids [2].Since that time much attention was devoted to the FIR spectra of the smallest of the amino acids,glycine [5–10],and alanine [9–14],while for other amino acids measure-ments of low-frequency vibrations have been reported only sporadically [4].No systematic study is available to comprise all of the 20a -amino https://www.wendangku.net/doc/c516217887.html,mon infra-red atlases do not present the complete far-infrared spectra [3],in general data below 400cm à1are missing.In this paper we present the absorbance spectra of 18amino acids in the frequency range between 10and

Corresponding author.Tel.:+497116854947;fax:+497116854886.

E-mail address:drichko@pi1.physik.uni-stuttgart.de (N.Drichko).1

On leave from Io?e Physico-Technical Institute,St.Petersburg,Russia.

D -L -alanine,L -cystine,L -leucine,

L -glutamic acid,

D -arginine (carbonate),D -L -proline,phenylalanine,tyrosine,

L -histidine

(monochlorhydrate).

https://www.wendangku.net/doc/c516217887.html,/locate/chemphys

Chemical Physics 316(2005)

61–71

650cmà1and discuss the features they have in common. Our assignment of the observed vibrational bands is based on the calculations of the normal modes available in literature and comparison of the spectra of di?erent amino acids.

2.Experimental procedure

The transmission spectra of the L-amino acids were obtained on pressed pellets of polycrystalline powders. The powders were purchased from Fluka and used with-out further puri?cation.The materials were stored according to the indications of the supplier.In order to limit scattering e?ects,the powder was ground prior to the pellets preparation,obtaining particles with a size of1–2l m.Each of the amino acids was mixed with polyethylene and pressed in disks of1cm diameter with a force of7kN.The thickness d of the resulting pellets was determined by a micrometer:typically between 200and400l m,for the low-frequency experiments even thicker.

Transmission measurements were performed using a modi?ed rapid-scan Fourier transform spectrometer Bruker IFS113v,equipped with a He-cooled bolo-meter.The measurements presented here cover the spectral region from10to650cmà1with a resolution of1cmà1.All the experiments were performed at 25°C.

Based on the recorded transmission spectra of the pellets(polycrystalline samples in polyethylene matrix), we calculated the absorbance A of the amino acids in the following way:

A?àloge1=TT;e1Twhere the transmission of the amino acid is given by: T?T S=T R;e2THere T S is the transmission measured through the pellet.The transmission of the bare polyethylene matrix of the pellet is given by:

T R?e1àR RT2expeàa RádácT.e3THere R R indicates the re?ectivity of the polyethylene, a R is the absorption coe?cient of polyethylene,d is the thickness of the pellet,and c indicates the concentration of the polyethylene in the pellet.The absorption coe?-cient of polyethylene was calculated as

a R?4p k m=10;e4Twhere k is the extinction coe?cient of polyethylene determined from?tting of the measured data.For the measurements in the range50–650cmà1we used pellets with the mass ratio amino acid/PET of1:10,while for the range10–50cmà1the ratio was1:1.3.Results and discussion

In Figs.1–5,the absorbance spectra of18amino

acids3are displayed;they are arranged in groups accord-ing to the structure of the molecules and to similarities of the spectra.On the right side of the spectra,the respective molecules are sketched.The frequencies of the absorption bands observed in the spectra are com-piled in Tables4–8.

Our discussion starts with the smallest amino acids, glycine and alanine,because an assignment of their spec-tra might help to understand those of the more compli-cated amino acids.It also allows us to compare our ?ndings with the literature[5–14]where FIR spectra of these two acids are reported and assignment suggested. On the other hand,in the present study we were able to detect low-frequency modes(lying below100cmà1) which were predicted by calculations but have not been observed in experiments yet.

The crystals of amino acids are molecular crystals, comparable weakly bound in the lattice by Wan der Waals forces,and thus one would expect that the spectra of the crystalline samples are dominated by the molecu-lar features.Indeed,without any exception the most intensive bands in the spectra belong to molecular vibra-tions;lattice vibrations are present only below100cmà1 and they have much lower intensity.The symmetry of the amino acids crystals and the number of intramolec-ular vibrations are summarized in Table1.

3.1.Glycine

Glycine is the smallest of the amino acids:its side group is a plain hydrogen atom.According to the 3Nà6rule(with N the number of atoms)it has24nor-mal modes:theoretically,all of them are infrared and Raman active[15].In its spectrum(Fig.1b)we observe four strong peaks between200and650cmà1.The small-est one at607cmà1is assigned as a CC@O bending[5], predicted by DFT calculations[16]to appear at 624cmà1.At504cmà1a strong COOàrocking/bend-ing/wagging mode is observed[5];it is asymmetric with a shoulder toward high frequency(548cmà1),which indicates an underlying NHt

torsion mode[17,18].It is interesting that the latter mode was not predicted by the latest calculations[15,16],although it evidenced in

earlier studies[17].However,we?nd that this NHt

tor-sion mode is present in the spectra of all amino acids.In

case of glycine,the NHt

torsion mode appears like a very weak shoulder,that develops into a strong peak when the temperature decreases[19].At356cmà1there

3The missing amino acids are cysteine and arginine.We could not provide safe conditions for working with them.

62 A.Matei et al./Chemical Physics316(2005)61–71

Fig.1.Absorbance spectra of glycine,alanine,leucine,and isoleucine:(a)region 10–50cm à1;(b)region 10–650cm à1

Fig.2.Absorbance spectra of glutamine,aspartic acid,glutamic acid and asparagine:(a)region 10–50cm à1;(b)region 10–650cm à1.

A.Matei et al./Chemical Physics 316(2005)61–7163

is a strong peak in the spectrum due to the CC a N bend-ing [17].

The far-infrared absorption of glycine is dominated by a broad band centered at 200cm à1which exhibits a series of peaks on its low-frequency wing.These excita-tions contain more than just one contribution:hydrogen bond deformations,skeletal modes,lattice vibrations,and other modes from the COO àand NH t3groups.In the region below 200cm à1we clearly identify weak

modes at 164and 134cm à1

(NH áááOC stretching [6],NH 3bending),and 87,77,68,and 59cm à1,which rep-resent hydrogen bond bending (NH áááOC),COO àtor-sions,CC a N deformations,NH t3torsions,in each case with a di?erent weight.Model calculations

presented

Fig.3.Absorbance spectra of histidine,tyrosine,phenylalanine,and tryptophan:(a)region 10–50cm à1;(b)region 10–650cm à1

Fig.4.Absorbance spectra of methionine,serine,and threonine:(a)region 10–50cm à1;(b)region 10–650cm à1.

64 A.Matei et al./Chemical Physics 316(2005)61–71

in[7]predict the lowest modes for glycine to be at40and 14cmà1in Raman,and29cmà1in infrared.Following experiments[20],there were made suppositions about the existence of another mode at33cmà1;there are no assignments for any of these low lying modes.Perform-ing a Lorentzian?t of the spectra,we were able to iden-tify modes at52.4,37,and at22cmà1.

3.2.Alanine

The second simple amino acid,alanine,crystallizes in an orthorhombic lattice with space group symmetry P212121[21];all the33fundamental internal vibrations of the molecule are expected to be both Raman and infrared active[22].Going from glycine to alanine the increasing number of modes becomes immediately obvi-ous:the resulting FIR spectrum is much more complicated.

In the alanine absorption spectrum,the COOàmodes are shifted to higher frequencies compared to glycine: the COOàrocking mode is found at647cmà1and the COOàrocking/bending/wagging mode is observed at 538cmà1(see also Table4).The NHt

torsion peak is located at486cmà1.Four peaks of medium to

strong Fig.5.Absorbance spectra of lysine,valine,and proline:(a)region10–50cmà1;(b)region10–650cmà1.

Table1

List of amino acids with their chemical formulas and the number of molecular vibrations according to3N-6

Amino acid Name Chemical formula Number of vibrations(3Nà6)Space group/Reference a-Glycine Gly(G)C2H5NO224P21/n[30]

L-Alanine Ala(A)C3H7NO233P212121[21]

L-Leucine Leu(L)C6H13NO260P21[31]

L-Isoleucine Ile(I)C6H13NO260P21[32]

L-Asparagine Asn(N)C4H8N2O348P212121[33]

L-Glutamine Gln(Q)C5H10N2O354P212121[34]

L-Aspartic acid Asp(D)C4H7NO442P21[35]

L-Glutamic acid Glu(E)C5H9NO451P212121[36]

L-Tyrosine Tyr(Y)C9H11NO366P212121[37]

L-Tryptophan Trp(W)C11H12N2O275Pmmm[38]

L-Histidine His(H)C6H9N3O254P21[39]

L-Phenylalanine Phe(F)C9H11NO263P2221[40]

L-Methionine Met(M)C5H11NO2S54P21[31]

L-Threonine Thr(T)C4H9NO345P212121[41]

L-Serine Ser(S)C3H7NO336P212121[42]

L-Proline Pro(P)C5H9NO245P212121[43]

L-Lysine Lys(K)C6H14N2O266Unknown

L-Valine Val(V)C5H11NO251P21[44]

L-Arginine Arg(R)C6H14N4O272Unknown

L-Cysteine Cys(C)C3H7NO2S36P21[45]

In the last column the space group symmetries of the amino acids crystals are listed.

A.Matei et al./Chemical Physics316(2005)61–7165

intensity are found between278and409cmà1;all of them have their origin in the CC a N deformation and CH3torsion[22,23].In the glycine spectrum the equiva-lent to those modes is the peak at357cmà1,which rep-resents only the CC a N bending.The bands of CO2 torsion at216and258cmà1are observed in alanine, but not in glycine spectra.Also the strongest low-fre-quency peak is shifted down to167cmà1in the alanine spectrum;nevertheless,the origin is the same.In the low-frequency range,the spectra of the two simplest amino acids exhibits similar features;all the peaks (112,104,85,73and60cmà1)have low intensity.These features are due to hydrogen-bond deformations cou-pled to other modes,like COOàtorsion,NH3torsion, and CC a N deformation.In alanine,CH3modes also contribute to the absorption below200cmà1.We were able to detect an absorption peak at60cmà1,which was not observed previously.This?nding is in agree-ment with DFT calculations[16]that predict the exis-tence of a band of NC a C torsion at62cmà1.

One would expect that for two amino acids with sim-ilar structure,the absorption spectra should be similar since they represent the characteristic vibrations of the molecules.This is in fact the case for the mid-infrared spectra of the amino acids.However,the situation is completely di?erent when it comes to the far-infrared frequency range.Our FIR absorption study reveals that the spectra of the amino acids are very particular(see Figs.1–5)and,as a matter of fact,they can serve as?n-gerprints for the molecules.We suggest a distinction of amino acids by FIR spectroscopy.On the other hand, some general features common to most amino acids may be extracted.Starting from the established assign-ment of the bands for glycine and alanine,we can now assign some bands in the spectra of the other amino acids,even if normal mode calculations have not been performed.

3.3.Leucine and isoleucine

Isoleucine and leucine are composed by adding one and two CH3groups to the side chain of alanine(Fig.

1).Their spectra contain clear similarities to the glycine and alanine spectra.The pronounced539cmà1peak in alanine becomes a doublet at about550cmà1(538–557cmà1)in the isoleucine spectrum and a band at 536cmà1in the case of leucine;accordingly,we assign them to COOàrocking/banding/wagging.In addition, the NH3vibration is clearly seen in the isoleucine spec-trum at490cmà1.The multiple bands in the300–400cmà1region belong to CC a N deformations and to CH3torsions.The low-lying strong bands of intramolec-ular vibrations shift to lower frequencies as the molecule size increases in this group of amino acids;they are found at142cmà1for isoleucine and at123cmà1for leucine.These absorption features are assigned to an overlap of hydrogen bond deformations,CC a N defor-mations,and lattice vibrations.

Table2

Tyrosine absorption frequencies:our experimental data compared with computed frequencies based on normal mode calculations by Grace et al.[28] m(cmà1)measured m(cmà1)calculated Mode

31.8w32.63Torsion of entire side chain about C-ring bond 52vw52.72Torsion of HOOCNH2HC-about chain C–C-bond 69w68.02C–H oop bend,in-phase(11)

88w87.76COOH torsion

111w––

116w––

131w––

160w163.20Amino torsion

171w–

–199.07Distal chain C–C–C bend

208w210.29Proximal chain C–C–C bend

248s––

–297.49O@C–C and H2N–C–C bends,in-phase

310s302.08Ring OH torsion

335w325.39C–H ip(trigonal)bend(15)

377s393.76N–C–C bend(w/proximal C–C-pair)

433m434.79C–H oop bend(C6libration)(10b)

473w461.25C–H ip bend(9b)

493m483.98C–C–C oop bend(16a)

529s518.73Carboxyl HO–C–C bend

575s559.94C–C–C ip bend(6a)

–610.14C–C–C oop bend(16b)

639vw,sh627.90Carboxyl OH torsion

650s––

66 A.Matei et al./Chemical Physics316(2005)61–71

3.4.Glutamic acid,aspartic acid,and related amino acids (Fig.2)

The overall-shape of the glutamic acid spectrum resembles that of glycine.It also consists of four wide bands,which fall in the same spectral regions:one at 100–250cm à1with a maximum at 227cm à1,the second one between 300and 450cm à1with a maximum at 376cm à1,and ?nally two bands located at 503and 536cm à1.The absorption bands observed in glutamic acid are in general wider and have some structure,in con-trast to the glycine spectrum.The low-lying band has the same origin as in glycine,i.e.it is governed by hydrogen bonds.Obviously there is much more structure on the low-frequency side.A very pronounced mode is observed at 41cm à1which is a COO àtorsion coupled with a CC torsion [24].Following the results from [25]we consider the wide band with a maximum at around 400cm à1as due to the backbone deformation,probably coupled with other molecular vibrations.According to our assignment above,these bands in the 500cm à1range belong

Fig.6.The vibrational peaks statistic in amino acids.

Table 4

Absorption frequencies for glycine,alanine,leucine,and isoleucine Gly Ala

Leu Ile Assignment

22w 21vw 10w Hydrogen bond modes

27w 28w 37vw 36w 52.4vw 43vw 49w 59w 59w 55w 58w

68w 71w

77w 73w 82w

87w

85w 104w 95w 112w

119w

123s

134w

142s 164w 167s 170w 175s 200s 205s 202vw 233

216s 258w 228w

238s COO àvibrations

278w 293w 294w CC a N deformations

313m

324m

333w 344s 343s 356s

364w 370w 409s

402s 393s NH t3modes 426w 444s

443w 486w

490

504s 539w

536s

538s COO àrock/bend/wag vibrations

548sh 557m

607s

647w

Table 3

Vibrational modes correlated with frequencies Frequency Vibration

Below 220cm à1Hydrogen bond modes 220–270cm à1COO àvibrations 270–380cm à1CC a N deformations 380–480cm à1NH t3modes

480–650cm à1

COO àrock/bend/wag vibrations

A.Matei et al./Chemical Physics 316(2005)61–7167

COO àrocking/bending/wagging;the excitation is split because there are two C @O groups in glutamic acid.It is not clear why the COO àbending mode around 600cm à1is missing in the case of glutamic acid.

The absorption spectrum of aspartic acid exhibits similar features:three well-de?ned bands which have a certain structure.We can safely assume that these wide bands have the same origin (Table 5).

Asparagine and glutamine present a common feature at about 540cm à1.This absorption peak visible in asparagine at 539cm à1and in glutamine at 541cm à1was by now assigned to a NH 2torsion [26].One of the reasons for such an assumption was that both amino acids have a NH 2group in the side chain.However,this feature appears in glutamic acid (536cm à1)and aspartic acid (553cm à1)as well,which makes us believe that we deal with a COO àvibrational mode,of the same type like in glycine [5].The NH t3torsion mode appears in these two amino acids at almost the same frequency [26,27]:455cm à1in glutamine,and 456cm à1in asparagine.At lower

frequencies the spectra become more speci?c for each molecule and only a few general features can be ex-tracted.As a rule,a strong absorbance maximum due to a backbone deformation,probably coupled with other molecular vibrations is present in the 300–400cm à1range.A wide intensive band with a maximum at about 200cm à1is also observed in all of the spectra.It is formed by an overlap of vibrations of H-bonds and complicated molecular vibrations involving COO àtorsion,NH 3tor-sion CC a N deformation.With increasing size of a mole-cule,the bands split into a larger number of small peaks.3.5.Phenylalanine,tyrosine and other amino acids with cyclic side groups (Fig.3)

This subgroup of amino acids includes the molecules,which have a phenyl ring as a part of the side group

Table 6

Absorption frequencies for histidine,tyrosine,tryptophan,phenylal-anine,and histidine His Tyr Phe

Trp Assignment

22.8w –

25.5s

31.8s

28w Torsion of entire side chain about C-ring bond

40.6w Hydrogen bond modes 52vw 61w

69w 69w 80w 88w 92w

74w 111w 129w

116w 122w

131w 139w

160w 159w

175w 171w

201s 195s

196s 215w

208w 215w 238m COO àvibrations 248s

252w 269w

307w 310s

CC a N deformations

315m 317w

325m

344s

335w 347s

377s

365s

397w NH t3modes

419s

426s

433m

469m

456w

473w

493m 499w COO àrock/bend/wag vibrations

509w 537s

529s

525s

530w 549w 559w 575s

581w 589

w 605w

626s

627w 639vw 654w

650s

655w

Table 5

Absorption frequencies for aspartic acid,glutamic acid,glutamine,and asparagine Asp

Glu Gln Asn Assignment 11w –

41w

COO àtorsion +CC torsion

57w

55s 57w CC torsion +COO àtorsion 67w 62w Hydrogen bond modes 82w 76w

87w

89w 84w

93w

CC torsion

100w 103w

Hydrogen bond modes

109w 108w 112w 119s 119w 132w 133w 140s 130s 148w 150w 159w 163w 177w 171s 174w 184w CCC bending +CC torsion 196w 209s 200s CCC bending +NH t3torsion 214s 227s

239w 224w 279w

253w 260w

COO à

vibrations

288s

CC a N deformations

313w 324w 345w 332s

360s 376s 353s

402

414m 416w

413w NH t3modes

449w

455w 456s

478w

503m COO àrock/bend/wag vibrations

553w 536s

541s

539m

581m 600w 657w

637w

68 A.Matei et al./Chemical Physics 316(2005)61–71

(Fig.3).The simplest of them is phenylalanine,it is based on alanine with one H-atom in a side group substituted by a phenyl group.Of course,the attachment of a big mass changes the vibrational spectra considerably.Nev-ertheless,the spectrum of phenylalanine is surprisingly simple:it exhibits only four peaks above 200cm à1and three signi?cant maxima in the spectral range below.Although the structure of phenylalanine and tyrosine are not so much di?erent,the FIR spectra change com-pletely.Tyrosine exhibits more than a dozen almost equally spaced absorption peaks.From the amino acids of this group,a normal-mode calculation was performed for tyrosine [28].In our study we observed the predicted bands;these calculations are also helpful in an assign-ment of less complicated phenylalanine spectrum.

In Table 2we compare the vibrational frequencies we have measured for tyrosine with the result of normal mode calculations [28].

In general,the computed values are very close to those observed;this is in particular the case for the comparably strong mode detected at very low frequencies (32cm à1)and assigned to the torsion of the entire side group [28].According to the calculations from [28],the 208cm à1peak in tyrosine spectra is assigned to a contribution of the C–C–C (backbone)bending the same assignment can be proposed for the strong absorption feature slightly below 200cm à1in phenylalanine spectra.

In the 300–400cm à1region tyrosine is richer in spec-tral features than phenylalanine.In the case of phenylal-anine,there is a prominent absorption peak at around

365cm à1,which in tyrosine corresponds to three peaks of lower intensity,at 310,335,and 377cm à1.Normal mode calculations [28]predict the positions of the respective absorbance bands:the torsion ring-OH bond at 302cm à1,the C–H in plane bend at 325cm à1,and the CC a N bending at 398cm à1;the agreement is very good.We assume that the peak observed in phenylala-nine is a CC a N bend overlapping with the in-plane bending of the chain-CH;spectroscopic studies on polyphenylalanine [29]prove an assignment of the 365cm à1band to the backbone bending vibration.The only vibration involving the phenyl ring seems to be given by the weak shoulders observed at 318cm à1in tyrosine and at 317cm à1in phenylalanine.According to the predictions from [28],we assign them to a phenyl-CH inplane bend,which is expected at 325cm à1.

The COO àgroup vibrations of phenylalanine are dis-placed toward lower frequencies compared to alanine (525and 605cm à1in phenylalanine,539,and 647cm à1in alanine).The same happens with the NH 3vibration:it shifts down in frequency to 469cm à1.

4.Conclusions

In the present work absorption spectra of 18amino acids were measured in the spectral region from 10to

Table 8

Absorption frequencies for proline,lysine,and valine Pro

Lys Val

Assignment

32.3w Hydrogen bond modes

47.5w

64w

55w 75w 95w 108w 125s 156w 145s 183s

205w 225s

246w

236s COO àvibrations

278w 295w 326m

336s

CC a N deformations

356s 394w 375s 417w 403w NH t3modes 454s

430s 428s 480w

471w 544s

COO àrock/bend/wag vibrations

551w

572w 644w

664w

Table 7

Absorption frequencies for methionine,threonine,and serine Met Thr Ser Assignment

37.4w

48w 38.6w Hydrogen bond modes

71w 64s 102m 140s

125w 148w 129w 165w 160w 191s 172w

203s

228s

COO àvibrations

251s

284m 314s 299s

CC a N deformations

335w 340m 364w 366s

378w 417s 417m

NH t3modes

448w 430w

447w 546s

491s 524m COO àrock/bend/wag vibrations

561s

564w 604m 639

643w 654w

A.Matei et al./Chemical Physics 316(2005)61–71

650cmà1.The assignment was based on calculations present in literature and comparison of the spectra of di?erent amino acids.For most of the amino acids the low frequency spectra were measured for the?rst time and the low frequency modes predicted by calculations were observed.Our results demonstrate that small changes in the structure of the organic molecules in?u-ence the spectra in the far-infrared range considerably, in contrast to the mid-infrared region.In fact,the far-infrared spectra are so speci?c for each amino acid that they can easily be distinguished with THz spectroscopy. On the other hand,we could identify and assign molec-ular vibrations by their similar positions in the observed spectra.The features are not as characteristic as in mid-infrared range,but a general trend can be extracted;it is presented in the diagram of Fig.6.The far-infrared spectra between10and650cmà1can be divided in seg-ments with a width of20cmà1.In this statistical ap-proach,the vertical bar indicates how many amino acids exhibit vibrational peak in this frequency window; the strength of the absorption was not taken into ac-count.Fig.6simply identi?es the frequencies regions in which it is more likely to have strong absorption in the spectrum of an amino acid.As the Table3indicates,?ve distinct frequency regions can be identi?ed and re-lated to certain types of vibrations:

Of course this label is not exclusive:there are regions in which di?erent types of vibration appear.For in-stance,the glycine peak around135cmà1results from

overlapping CC a N torsion,NHt

3bending,and hydro-

gen bond(HO)stretching.Although the region below 220cmà1belongs to the hydrogen-bond modes,other vibrations appear here as well:backbone deformations,

and the vibrations of the end groups NHt

3and COOà.It

is still an open question where the lower frequency limit for H bonds is.Around270–380cmà1the most promi-nent vibrations are the CC a N deformations.The NHt

3 modes appear around380–480cmà1.The COOàmodes are frequently found in the spectra above480cmà1as well as between220and270cmà1.Low-frequency vibrations are more delocalized,i.e.more atoms contrib-ute to these vibrations.We can note from the diagram in Fig.6that at low frequencies,the bands become less characteristic and their positions are more speci?c for each amino acid.Indeed,the low-energy excitations are stronger in?uenced by di?erences in the molecular and crystal structure.

Acknowledgements

The project is funded by the EU under Contract QLK4-CT2000-00129THz-Bridge).One of us(N.D.) enjoys support of the Alexander von Humboldt-Foundation.References

[1]A.Barth,Prog.Biophys.Mol.Biol.74(2000)141.

[2]E.Heintz,Compt.Rend.Hebd.Seances Acad.Sci.201(1935)

1478.

[3]D.Schrader,Raman,Infrared Atlas of Organic Compunds,

Verlag Chemie,Weinheim,1989.

[4]G.Dovbeshko,L.Berezhinsky,J.Mol.Struct.450(1998)121.

[5]M.Tsuboi,T.Onishi,I.Nakagawa,T.Shimanouchi,S.Mizu-

shima,Spectrochim.Acta12(1958)253.

[6]S.Suzuki,T.Shimanouchi,M.Tsuboi,Spectrochim.Acta19

(1963)1195.

[7]K.Machida, A.Kagayama,Y.Saito,Y.Kuroda,T.Uno,

Spectrochim.Acta A33(1977)569.

[8]J.B.Hasted,S.K.Husain,A.Y.Ko,D.Rosen,Elisabeth Nicol,

in:H.Fro¨hlich, F.Kremer(Eds.),Coherent Excitations in Biological Systems,Springer Verlag,Berlin,Heidelberg,1983, pp.71–83.

[9]L.Genzel,F.Kremer,A.Poglitsch,G.Bechtold,in:H.Fro¨hlich,

F.Kremer(Eds.),Coherent Excitations in Biological Systems,

Springer Verlag,Berlin,Heidelberg,1983,pp.58–70.

[10]N.Wright,J.Biol.Chem.120(1937)641;

N.Wright,J.Biol.Chem.127(1939)137.

[11]K.Fukushima,T.Onishi,T.Shimanouchi,S.I.Mizushima,

Spectrochim.Acta(1959)236.

[12]S.Suzuki,T.Ohshima,N.Tamiya,K.Fukushima,T.Shima-

nouchi,S.I.Mizushima,Spectrochim.Acta11(1959)969. [13]J.Bandekar,L.Genzel,F.Kremer,L.Santo,Spectrochim.Acta

A39(1983)357.

[14]K.Machida,A.Kagayama,Y.Saito,T.Uno,Spectrochim.Acta

A34(1978)909.

[15]A.Vijay,D.N.Sathyanarayana,J.Phys.Chem.96(1992)10735.

[16]S.G.Stepanian,I.D.Reva,E.D.Radchenko,M.T.S.Rosado,

M.L.T.S.Duarte,R.Fausto,L.Adamowicz,J.Phys.Chem.A 102(1998)1041.

[17]C.Destrade,C.Garrigou-Lagrange,M.-T.Forel,J.Mol.Struct.

10(1971)203;

B.Dupuy,These,Bordeaux,https://www.wendangku.net/doc/c516217887.html,RS AO1275,1967.

[18]S.E.Kettle,E.Lugwisha,J.Eckert,N.K.McGuire,Spectrochim.

Acta A45(1989)533.

[19]S.K.Husain,J.B.Hasted, D.Rosen, E.Nicol,J.R.Birch,

Infrared Phys.24(1984)201.

[20]B.M.Powell,P.Martel,Chem.Phys.Lett.67(1979)165.

[21]H.J.Simpson Jr.,R.E.Marsh,Acta Cryst.20(1966)550.

[22]C.H.Wang,R.D.Storms,J.Chem.Phys.55(1971)3291.

[23]M.Barthes,A.F.Vik,A.Spire,N.N.Bordallo,J.Eckert,J.Phys.

Chem.A106(2002)5230.

[24]F.J.Ramirez,J.T.Lopez Navarrete,Spectrochim.Acta2(1995)

293.

[25]J.T.L.Navarrete,V.Hernandez,F.J.Ramirez,J.Raman Spectr.

25(1994)861.

[26]J.Casado,J.T.Lopez Navarrete,F.J.Ramirez,J.Raman Spectr.

26(1995)1003.

[27]P.Dhamelincourt,F.J.Ramirez,Appl.Spectr.47(1993)446.

[28]L.I.Grace,R.Cohen,T.M.Dunn,D.M.Lubman,M.S.deVries,

J.Mol.Spectr.215(2002)204;

M.J.Frisch,G.W.Trucks,M.Head-Gordon,P.Gill,M.W.Wong, Gaussian92,Revision C,Gaussian Inc.,Pittsburg,PA,1992. [29]A.Xie,Q.He,https://www.wendangku.net/doc/c516217887.html,ler,B.Sclavi,M.R.Chance,Biopolymers49

(1999)591.

[30]G.Albrecht,R.B.Corey,J.Am.Chem.Soc.61(1939)1087.

[31]B.Khawas,Acta Cryst.B26(1970)1919.

[32]K.Torii,Y.Iitaka,Acta Cryst.B27(1971)2237.

[33]J.J.Verbist,M.S.Lehmann,T.F.Koetzle,W.C.Hamilton,Acta

Crystallogr.Cryst.Chem.28(1972)3006.

[34]W.Cochran,B.R.Penfold,Acta Cryst.5(1952)644.

70 A.Matei et al./Chemical Physics316(2005)61–71

[35]J.L.Derissen,H.J.Endeman,A.F.Peerdeman,Acta Cryst.B24

(1968)1349.

[36]S.Hirokawa,Acta Cryst.8(1955)637.

[37]R.Boggs,J.Donohue,Acta Cryst.B27(1971)247.

[38]B.Khawas,G.S.R.Krishna Murti,Acta Cryst.B25(1969)1006.

[39]J.J.Madde,E.L.McGandy,N.C.Seeman,M.M.Harding,A.

Hoy,Acta Cryst.B28(1972)2382.

[40]B.Khawas,Indian.J.Phys.A59(1985)219.[41]B.L.Silva,P.T.C.Freire,F.E.A.Melo,I.Guedes,M.A.Araujo

Silva,J.Mendes Filho,A.J.D.Moreno,Braz.J.Phys.28(1998)

19.

[42]T.J.Kistenmacher,G.A.Rand,R.E.Marsh,Acta Cryst.B30

(1974)2573.

[43]B.A.Wright,P.A.Cole,Acta Cryst.2(1949)129.

[44]K.Torii,Y.Iitaka,Acta Cryst.B26(1970)1317.

[45]M.J.Harding,H.A.Long,Acta Cryst.B24(1968)1096.

A.Matei et al./Chemical Physics316(2005)61–7171