Anomalous vibrational modes in acetanilide as studied by inelastic neutron scattering

HAL Id: jpa-00246673

https://hal.archives-ouvertes.fr/jpa-00246673

Submitted on 1 Jan 1992

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Anomalous vibrational modes in acetanilide as studied by inelastic neutron scattering

Mariette Barthes, Juegen Eckert, Susanna Johnson, Jacques Moret, Basil Swanson, Clifford Unkefer

To cite this version:

Mariette Barthes, Juegen Eckert, Susanna Johnson, Jacques Moret, Basil Swanson, et al.. Anomalous

vibrational modes in acetanilide as studied by inelastic neutron scattering. Journal de Physique I,

EDP Sciences, 1992, 2 (10), pp.1929-1939. �10.1051/jp1:1992256�. �jpa-00246673�

J.

Fhys.

Ifrance 2

(1992)

1929-1939 _OCTOBER t992, PAGE 1929

Classification

Physics

Abstracts 34.30 34.50

Anomalous vibrational modes in acetanilide

_as

studied by

inelastic neutron scattering

Mariette Barthes

(I), Juergen

Eckert

(2),

Susanna W. Johnson

(2), Jacques

Moret

(I),

Basil I. Swanson

(3)

and Clifford J. Unkefer

(3)

(~)

Groupe

de

Dynarnique

des Phases Condensdes, Universit£ des Sciences, 34095

Montpellier,

France

(~) LANSCE, Los Alamos National Laboratory, Los Alamos, NM 87545, U.S.A.

(~) INC4, Los Alamos National

Laboratory,

Los Alamos, NM 87545, U-S-A- (Received 19 November 1991,

accepted

in

final form

19 June 1992)

Abstract. A

study

^{of the} anomalous modes in acetanilide and five deuterated derivatives by incoherent inelastic neutron scattering is

reported.

These data show that the

dynamics

of the amide and methyl groups influence each other. In addition, the anomalous temperature behaviour of the NH out-of-plane bending ^{mode is} confirmed. These observations suggest ^{that the} self-trapping

mechanism in ACN may be _more

complex

than hitherto assumed.

The

origin

of the anomalous infrared and Raman modes in

crystalline

acetanilide

(C~HSNHCOCH~,

_or

ACN) II,

remains _a

subject

of considerable controversy. ^One

family

of theoretical models involves

Davydov-like

solitons

[2],

nonlinear vibrational

coupling [3]

_or

«

polaronic

_» localized modes

[4, 5].

An altemative

interpretation

of the extra-bands in _terms of _a Fermi _{resonance was}

proposed [6]

and

recently

the existence of

slightly non-degenerate hydrogen

atom

configurations [7]

in the H-bond was

suggested

as an

explanation

for the

anomalies.

Acetanilide, displaying

chains of

hydrogen

bonded amide groups, is _a model system ^for

alpha-helix proteins.

Its

study

could shed _some

light

_on the unknown mechanism of energy transfer in

biological

molecules

[8].

In this paper we report some new results _on the anomalous vibrational modes in ACN that

were obtained

by

inelastic incoherent neutron

scattering (INS). Comparison

of the spectra ^of ACN and of five deuterated derivatives

greatly

facilitates

assignments

of the main features.

The mutual influence of the different chemical groups

(amide, methyl

_group and

phenyl ring)

_are

emphasized by

means of selective deuterations. The

possibility

of vibrational

coupling

between them is discussed. In addition, the temperature

dependence

of the

~NH)

modes

(out-of-plane

bend of the

H-bond)

and of the

methyl

torsional modes _are described.

These data _are then

compared

to the

predictions

of the different theoretical models that attempt

to

explain

the anomalies.

1930 JOURNAL DE PHYSIQUE ^{I N° lo}

The INS data _were collected _on the Filter Difference

Spectrometer (FDS),

at Manuel

Lujan

Jr. Neutron

Scattering

Center of Los Alamos National

Laboratory,

between lo K and 300 K.

The useful range of energy transfers _was loo cm-I-1600 cm-1

(l2meV-200meV)

with _a

relative resolution

(AE/E)

of about 2 fb.

The data measured with this instrument _{are a} convolution of the

scattering

law of interest with _a resolution function. The deconvolution

procedure

^{is achieved} with the aid of either of two

algorithms [9],

the Mezei method _{or a} Maxent reconstruction _to determine the

frequency

distribution.

The present

study complements

_a

previous investigation

with IN6 at the Institut Laue-

Langevin [10],

in which the range of transferred energy was about 0-400 cm-1

(50 mev).

Moreover,

FDS and IN6 do not cover the _same temperature range.

The

syntheses

of

powdered samples

_were carried out in _our laboratories. The

degree

^of

deuteration of the different molecular groups were monitored

by

IR and Raman, and _{was more} than 90 fb for the proton ^{of the amide} group, ^{and better than 99 fb for the other} groups.

Samples

were

always

handled in

dry

helium

atmosphere.

The five deuterated derivatives _are

C6HSNHCOCD3 ACN-methyl d3 C~H~NDCOCD3 ACN-methyl

d~ ^N-d _;

C~D~NHCOCH~. ACN-phenyl dsl C6D~NDCOCH~. ACN-phenyl

d~ ^N-d ;

C6D~ND- COCD~ ACN-d~.

Comparison

of the INS

spectra

^{of these various ACN}

isotopomers (or isotopic isomers)

is _a

powerful

tool for

determining

the

origin

of the modes and their

assignments

the

scattering

cross-section for H is _more than _an order of

magnitude greater

^{than that of D.} Deuteration therefore _not

only

shifts lines

(as

in

optical spectroscopy)

but also has _a drastic effect _on the

intensity

of the

respective

bands.

The low

temperature

spectrum ^of ACN is

displayed

in

figure

I from 100 to 800 cm-'.

Figure

2 shows the INS spectrum ^{in the}

region

^of

~NH)

at 12 K for the three

samples

that have

a

protonated

amide group. Tentative

assignments

of the bands in this

region

are

given

in table I. These _are in agreement ^with available

previous

data from Raman and IR spectroscopy.

However,

the INS

spectra

^do

provide

new information.

A rather

significant portion

of the

intensity

of _some bands in the INS spectra ^{of ACN} appears to be related to some

coupling

or combination modes that involve the

methyl

_group. This effect is demonstrated

by plotting

^{the difference}

spectra

: for

example (ACN-ACN (methyl d~)),

which should

only

leave lines of

methyl

_group modes, reveals that for _some

phenyl

and amide

modes,

_some

intensity

remains after the subtraction.

Similar effects _were

previously reported [I I]

and

explained by

a strong

conjugation

between amide and

methyl

_groups

(C~C~

distance is m1.48

h)

_{whose motions}

are less

independent

than

they

would be if the

C~CB

link _{were a}

single

^bond

(m1.54 h).

Figure

3 shows the

methyl

torsional modes at 12 K in three

samples.

In

spite

of _a limited

resolution,

the

methyl

^mode _appears to be

split,

in

agreement

^{with the Raman}

scattering spectra [12, 13].

In the

single crystal polarized

Raman

study [13]

with _a

z(~y

z

configuration

the

methyl

torsion _was found _to

gradually split

as the temperature ^is lowered below about 200 K. The

frequencies

of the three components as well _as the

splittings

all increase with

decreasing temperature.

^{INS studies}

[10] previously

identified the

methyl

torsion in ACN but the

splitting

_was not observed. In the present ^work at least three components may be identified in

ACN, namely

at

142,

146 and 152 cm-I. The effect of deuteration of the

phenyl ring

is _a

significant change

in the

shape

and width of the

methyl

librational band. In

ACN-phenyl d~-Nd

the

splitting

is _{even more}

evident,

and the relative intensities of the

components

^have

changed.

These observations demonstrate the influence _on the

dynamics

of the amide and

methyl

_groups

on each

other,

_or at least, the

sensitivity

of the

methyl

^torsion to its local steric environment.

The

frequency

shift of the

methyl

torsions _{as a} function of temperature ^{obtained from} our

data is shown in

figure

4 and is consistent with the measurements of Johnston et al.

[13].

N° IO ANOMALOUS MODES OF ACETANILIDE _: NEUTRON SCATTERING 1931

Table I. Tentative

assignment of

the INS spectrum

of

acetanilide.

Assignment

Observations

142-146-152

Methyl

torsions

Already assigned

in

[10]

and

[13]

Split,

sensitive _to the deuteration of other groups

186

Presumably

C-N torsion

Displays progressive

_energy shift with selective deuteration

275

Phenyl

modes

345

Methyl

modes

406

(C-C-C) out-of-plane

deformation

502

Methyl

modes

521

(C-C-C) out-of-plane

deformation

600

Methyl

modes

646

Methyl

683

Phenyl

modes, C-C-H deform.

754

~NH) (out-of-plane bending

Anomalous increase of its

mode) intensity

with

decreasing

temperature

[10, 14]

~CH), Phenyl

829

~CH), Phenyl

890 Combination band Decreases in

ACN-methyl

d~

and in

ACN-phenyl

d~

959

~CH) Phenyl Breathing

modes

[I]

020

Methyl

Rock

140

3~CH), Phenyl

Note _: Vibrational modes that involve

predominantly methyl

_group motions appear ^to be

heavily coupled

with other modes (see

Fig.

3). Their

precise assignment requires

_a norrnal coordinate

analysis.

_y

and 3 refer, to out-of-plane and

in-plane

bends,

respectively.

1932 JOURNAL DE PHYSIQUE I N° IO

%

ACN, 150 K

~

~

~~

~

= W

I in

O

(

o

200 400 600

Wave Number

cm"~

(a)

ACN, 12 K

$

~~_O

CT

w ~

(

o

200 400 600

Wave Number (

cm"~

(b)

Fig.

I. INS spectra ^of ACN at 150 K and 12 K.

The

~NH)

^mode at 754 _cm- is also found to be affected

by

deuteration of the other groups

as shown in

figure

2 in which the data from

ACN, ACN-methyl

d~ and

ACN-phenyl d~

are

displayed.

This mode

(also

shown in

Fig. 5a) undergoes

_a

change

in

shape

and width upon deuteration of the

methyl

_or

phenyl

_group. Similar

changes

in

shape

were observed in the

corresponding

IR

absorption

^bands

[14, 15].

This is shown

by comparing figure

5b with

figure

6b.

N° IO ANOMALOUS MODES OF ACETANILIDE _: NEUTRON SCATTERING 1933

~

12K ACN

#

$

~ B

C

fiw

~ O

§

m

o

700 800 900 1000

«

3

_{12K ACN}

(methyl-d~)

~

I

~w

~

c ^x

3 "

o

700 800 900 1000

~o 14K ACN

(pheny'-d~)

T

~l

i

~

~

C

$

3

o

700 800 900 1000

£(

Cm~

Fig.

2. INS spectra ^{of the} I~NH) mode in ACN,

ACN-methyl

d~,

ACN-phenyl

d5.

On the contrary, the bands associated with

phenyl ring

motions do not show such effects upon deuteration of the amide _or the

methyl

_group. The

only anomaly

is the contribution _to the

intensity

of these bands from the presence of the

methyl

_group.

Another

important

feature of these

spectra

^{is the}

temperature dependence

of the

~NH)

mode,

at 750-770 cm-I The

intensity

of this mode _was observed _to increase

strongly

with

1934 JOURNAL DE

PHYSIQUE

I N° lo

~g

^{ACN 12K}

i

~

ij

~O

( ~

U'

o

ioo iso ( cm- )

~ ACN

(pheny'-d~)

^14K

~

i (

m

o

ioo iso

cm-~)

~

ACN

(phenyl-d~,

^{N-d 12K}

$

i~

~ 5

C

(

w ~O

~

o

ioo iso E cm-

Fig.

3.

Methyl

torsional modes in ACN,

ACN-phenyl

d5,

ACN-phenyl

d5 Nd.

decreasing temperature (Fig. 5a). However,

_some

part

^{of this}

intensity

_appears to be the result of _some interaction with the

methyl

_{group, as we} described above, This

portion

_may be

roughly

estimated _to be about 30 fb of the total

intensity

of this band,

It is therefore _not clear whether the apparent ^{increase in}

intensity

with

decreasing

N° Io ANOMALOUS MODES OF ACETANILIDE NEUTRON SCATTERING 1935

- 150

5

I

~

z 140

#

~

>

130

~

§~

i

120

100 200 300 T (K

Fig.

4.

Energy

shift of the

methyl

torsional modes with respect to temperature : (.) ACN _; (6) ACN-

phenyl

d5 (x)

ACN-phenyl

d5 Nd,

temperature ^is an intrinsic

property

^{of the}

~NH)

mode or is related _to the temperature evolution of librational modes of the

methyl

_group. In order _to estimate the intrinsic evolution of the

intensity

of

~NH)

with temperature, one considers the two

spectra

^of

ACN-methyl d3

shown in

figure 6a,

in which there is _no

longer

_any contribution from the

methyl

_{group :} it is

apparent

^{that the}

intensity

of the

~NH) peak

increases much _more than other

peaks

with

decreasing temperature.

^Such an effect is consistent with infrared data

(Fig, 6b), Therefore,

the unusual

intensity

increase of the

~NH)

with

decreasing

temperature ^is an intrinsic

property

^of this mode.

These observations _are in

agreement

^{with IR}

spectra [14, 15]

which indicate _a total increase of

intensity

of the

y-NH

with

decreasing temperature

^{of about 20 fb between 300 K and 15 K,}

Similar observations have also been made

by

Raman

scattering

in ACN

[16],

Data from the different

techniques

are thus in agreement ^{about this}

important

result,

namely

that

~NH)

also has _an anomalous temperature

dependence.

Thus, ^ACN exhibits not

only

_an

anomalous amide mode at 1650 cm-~ _but also another _{one at} about 750 cm-I,

Among

these _new

results,

two of them _{seem to} be

especially

relevant _to the

problem

of the

anomalous modes in ACN _:

I)

the band

corresponding

to the

methyl

torsional transitions is

split

into three

components,

and the relative

intensity

of each component ^{is sensitive} to the deuteration of other parts ^{of the}

molecule _;

2)

there is _an anomalous increase of the

integrated intensity

of the N-H

out-of-plane bending

mode _at low

temperatures.

Considering

first the

splitting (also

observed in the

polarized

Raman

scattering [13])

of the

methyl librations,

the

theory predicts

that the energy levels of the

methyl

rotator submitted _{to a} rotational

potential

^{of threefold}

symmetry

occur in groups of three _a

degenerate pair

labelled

E~~, E~~

and _a

singlet A~ [17].

The

splitting

between A and E levels results from the

tunneling

splitting

of the torsional level _a. Then _two components are

expected

from the

theory,

and not three _as observed, A

lifting

of the

degeneracy

of the E levels is however

predicted

when the

methyl

libration

couples

with _a lattice mode

[17],

This

splitting

_may also be rationalized

by

a small

departure

^{from three-fold} symmetry ^{for the} rotational

potential.

This could be induced

by

the low symmetry ^{of the steric} environment of the

methyl

_group which in _tum could be caused

by

differences in the

positions

of the

hydrogen

bond protons, as

proposed by

Fann _et al,

[7],

The

methyl

_group would _{as a} consequence occupy

energetically inequivalent positions

and therefore have different librational

frequencies,

1936 JOURNAL DE PHYSIQUE I N° 10

600 700 800 700 600 700 BOO

io~ io~

250 K 175 K loo K

j

^{0.Sx10~ 10~}

*i

12K

~~~°~

~

i~

io~

£

w ₂₀₀ 150 K 55 K

io~

io~

600 Boo

600 800 700 600 BOO

la)

WAVENUMBERS cm~~

ACN 20K

300 K

it

<

~

w

#

g

,,

#

700 750 800

b)

WAVENUMBERS cm"~

Fig. 5.

Temperature dependence

of the ~NH) mode in ACN _: a) INS spectra ; b) Infrared absorption from [14, 15].

The

change

in

shape

and width of the librational components ^{when the amide} or

phenyl ring

_are

deuterated could result either from _a direct vibrational

coupling

of the

methyl

torsion with motions of the other groups

(or

with low

frequency phonons)

_or from _a modification of the

rotational

potential by changing

the local environment of

CH~.

The former model is consistent with the

hypothesis

of _«

polaronic

_» local modes _or solitons in ACN

[5] assuming

that the

methyl

torsion is

nonlinearly coupled

to the amide mode

-,

while the latter would be related to the

assumption

of

multiple

conformations of the molecular chain

[7].

N° 10 ANOMALOUS MODES OF ACETANILIDE _: NEUTRON SCATTERING 1937

ACN

methyl-d~

#

'~ ^{~'O 1'}

< ~ ^r

'

_

'

~

~°

/

^{,-, look}

c

j '

' '

~'

1 ^'

,-,'

~ ~' l ^'

,'~' ^'

" ° j

'-'

fi & ~

w ,-'

' '

~

'

o '

X l~l

~~~ i'

o "

700

j '/.

l

O ^'

~

'-

~

.

_Cl

~

~' ~~O

~

@ Cl

@

w -

c X

Tto 700

Fig.

estimated ^ackground) ; b) _absorption 5].

1938 JOURNAL DE PHYSIQUE I N° 10

m

o

O loo

T (K)

Fig. 7.

data

ACN)

(++) FDS

data(ACN

phenyl

d5) ; _(mm) FDS data

phenyl

d~

(ACN) ; ) _{calculated scattering}

cross-section.

The d changes

of

the

deuteration of other

groups of

the ^olecule could

be

for by

the

same

mechanisms,

I-e- either direct vibrational ^oupling or a multiple-well^otential for the mide

proton. In

the

latter case, changes in

the

steric

affects the

shape of this potential and thus

frequencies and

intensities

^{of the} transitions to the

excited

levels.

At

this point it is not ossible to decide which

Thesecond

result,

hichhas now

been

bserved by INS, _IR, andRaman

scattering, is

_the

anomalous thermal _behaviour of _{~NH), namely} an increase

of the

ntegrated intensity with

decreasing temperature. It _may

be expected

to provide urther mportant

input into

the

determination

of the origin of

these For

example,

in the

case of

(polarons, olitons or coupled scillators) the temperature ependence

of

the nomalous

intensity

should

obey

a characteristic

law I

(T),~o~

m exp (-

non-degenerate substates

for

_the amide _proton^uld

mean

either a ^mperatureindependent

global

intensity, or one govemed

by

^the Boltzmann population _of each

level.

our former

infrared data

[14]

ndicate that the

intensity

^of the ~NH)

exp(- T~/&~) law which_would favor the _family

of

odels of localized nonlinear^citations.

However,

former

_theories of the ACN

problem

only

_took into ccount the anomaly at

1650

cm-', and the _recent

observations

of new

_anomalies, for example in

the

hermal

behaviour

of

the _{~NH) mode,} suggest that the

self-trapping

echanism

in

_ACN may be

complex. urther

analysis

is in

_progress.

Acknowledgments.

This

Center, a _national user facility _{funded as such} by theepartment of Energy, _Office ^of

Basic Energy

This _work

is

supported _by NATO ^nder grant

n°910281,

and

by

the

cooperation CNRS/NSF.

N° 10 ANOMALOUS MODES OF ACETANILIDE NEUTRON SCATTERING 1939

References

II CARERI G., BUONTEMPO U., GALLUzI F., SCOTT A. C., GRATTON E, and SHYAMSUNDER E.,

Phys.

Rev. B 30 (1984) 4689.

[2] EILBECK J. C., LOMDAHL P. S. and SCOTT A. C.,

Phys.

Rev. B 30 (1984) 4703.

[3] TAKENO S.,

Frog.

Theor.

Phys.

75

(1986)

1.

[4] ALEXANDER D. M, and KRUMHANSL J. A.,

Phys.

Rev. B 33 (1986) 7172.

[5] SCOTT A. C., BIGIO I. J. and JOHNSTON C. T.,

Phys.

Rev. B 39 (1989) 12883.

[6] JOHNSTON C. T., SWANSON B. J., Chem.

Phys.

Lett. 114 (1985) 547.

[7] FANN W., ROTHBERG L., ROBERSON M., BENSON S., MADEY J., ETEMAD S. and AUSTIN R., Phys.

Rev. Lett. 64 (1990) 607.

[8] Cf, for example

Davydov's

Soliton Revisited,

Self-Trapping

of Vibrational Energy ⁱⁿ Protein, P. Christiansen and A. C. Scott Eds. (Plenum, N-Y-, 1990).

[9] SIVIA D., VORDERWISCH P. and R. SILVER, Nucl. Instrum. Meth. A 290 (1990) 492.

[10] BARTHES M., ALMAIRAC R., SAUVAJOL J. L., MORET J., CURRAT R., DIANOUX J., Phys. Rev. B 43 (1991) 5223.

[I II ABBOTT N. and ELLIOTT A., Proc. Roy. Sac. London A 234 (1956) 247.

[12] SAUVAJOL J. L., MORET J., ALMAIRAC R. and BARTHES M., J. Ram. Spect. 20 (1989) 517.

[13J JOHNSTON C. T., SWANSON B., ECKERT J, and UNKEFER C. J., J.

Phys.

Chem. (in

press).

[14] BARTHES M., in

Self-Trapping

of Vibrational Energy in Protein,

Davydov's

Soliton Revisited, P.

Christiansen and A. C. Scott Ed. (Plenum Press, 1990).

[15] KELLOUAI H. (thesis,

unpublished).

[16] JOHNSTON C. T, and SWANSON B. I.

(unpublished).

[17] CLOUGH S., ALEFELD B, and SUCK J. B., Phys. Rev. Lett. 41 (1978) 124.

[18] MARSHALL W. and LOVESEY S., Theory of Thermal Neutron Scattering (Oxford 1971) Chap. 4.