Relating structure with morphology: A comparative study of perfect Langmuir-Blodgett multilayers

Article abstract of DOI:10.1016/j.cplett.2007.11.069

Atomic force microscopy and X-ray reflectivity of metal-stearate (MSt) Langmuir-Blodgett films on hydrophilic Silicon (1 0 0), show dramatic reduction in 'pinhole' defects when metal M is changed from Cd to Co, along with excellent periodicity in multilayer, with hydrocarbon tails tilted 9.6° from vertical for CoSt (untilted for CdSt). Near edge X-ray absorption fine structure (NEXAFS) and Fourier transform infra-red (FTIR) spectroscopies indicate bidentate bridging metal-carboxylate coordination in CoSt (unidentate in CdSt), underscoring role of headgroup structure in determining morphology. FTIR studies also show increased packing density in CoSt, consistent with increased coverage.

Full text of DOI:10.1016/j.cplett.2007.11.069

Available online at www.sciencedirect.com
Chemical Physics Letters 451 (2008) 80–87
www.elsevier.com/locate/cplett
Relating structure with morphology: A comparative study of
perfect Langmuir–Blodgett multilayers
a
a,
b
b
*
Smita Mukherjee , Alokmay Datta , Angelo Giglia , Nichole Mahne ,
Stefano Nannarone b,c
a Surface Physics Division, Saha Institute of Nuclear Physics, 1/AF Bidhannagar, Kolkata 700 064, India
b TASC-INFM National Laboratory, Area Science Park, Basovizza (Trieste), Italy
c INFM and Dipartimento di Ingegneria dei Materiali ed Amb., Universit di Modena e Reggio Emilia, Italy
Received 15 October 2007; in final form 24 November 2007
Abstract
Atomic force microscopy and X-ray reflectivity of metal-stearate (MSt) Langmuir–Blodgett films on hydrophilic Silicon (100), show
dramatic reduction in ‘pinhole’ defects when metal M is changed from Cd to Co, along with excellent periodicity in multilayer, with
hydrocarbon tails tilted 9.6° from vertical for CoSt (untilted for CdSt). Near edge X-ray absorption fine structure (NEXAFS) and
Fourier transform infra-red (FTIR) spectroscopies indicate bidentate bridging metal-carboxylate coordination in CoSt (unidentate in
CdSt), underscoring role of headgroup structure in determining morphology. FTIR studies also show increased packing density in CoSt,
consistent with increased coverage.
Ó 2007 Elsevier B.V. All rights reserved.
1. Introduction
of such films are far from understood. A possible way of
reducing pinholes involve change in subphase pH [14,15].
However, high subphase pH increases out-of-plane fluctua-
tions in the multilayer [15] and slight variations in deposition
conditions seriously impede the deposition process [14]. An
alternative way involves subphase metal ions. Structure of
Langmuir monolayer is known to vary with subphase metal
ions [16,17]. Moreover, collapse behavior of stearic acid
monolayers with different metal ions [18] has shown that,
beyond monolayer collapse pressure, Co ions in aqueous
subphase give rise to coherent growth of multilayers incor-
porating the metal on water surface, while Cd ions lead to
multilayer islands of random size and height.
Here we have utilized this clue from collapse behavior to
reduce ‘pinhole’ defects in metal stearate LB films by intro-
ducing Co ions in subphase instead of Cd under same
deposition conditions. Co is magnetic in nature and nano-
particles of these and their compounds are of immense
interest in application oriented research [19]. Perfect LB
multilayers grown by stacking of consistently well ordered
defect-free layers are potential templates for growth of well
Metal-organic multilayers such as Langmuir–Blodgett
films [1] have potential applicability in molecular electronics,
non-linear optics, cell membrane models, biosensors and as
templates for growing functional nanomaterials like quan-
tum dots [2–10]. LB deposition process is a surprisingly easy
and economical way to build well-defined multilayers, from
Langmuir monolayers of amphiphilic molecules on water,
with molecular-level control [11] comparable with sophisti-
cated and expensive techniques like molecular beam epitaxy
(MBE) or metal-organic chemical vapor deposition (MOC-
VD). In spite of these appealing features, device applications
of these fail due to low reproducibility owing to presence of
‘pinhole’-type growth-defects [12], which are even treated as
‘characteristic features’ of LB growth [13]. Moreover, exact
interactions and bonding involved in headgroup structure
*
Corresponding author. Fax: +91 033 2337 4637.
E-mail address: alokmay.datta@saha.ac.in (A. Datta).
0009-2614/$ - see front matter Ó 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.cplett.2007.11.069

S. Mukherjee et al. / Chemical Physics Letters 451 (2008) 80–87
81
ordered magnetic arrays comprising nanoparticles of
Cobalt.
recorded by total electron yield (TEY). They were normal-
ized to an I0 spectrum previously measured for clean cop-
per to account for oxygen in mesh. Monochromator slits
ð50 lmÞ were set to provide an energy resolution of 0.1 eV.
Present work is a comparative study of Co-bearing LB
multilayers with Cd-bearing ones. Reciprocal space X-ray
reflectivity (XRR) is complemented with real space atomic
force microscopy (AFM) to study observed morphological
differences in CoSt and CdSt LB films. Goodness of multi-
layer deposition has been parameterized by two factors v
and v0. Information about metal ion-headgroup attach-
ment and hydrocarbon tail packing have been obtained
from FTIR spectroscopy. Local structure of CO bonds in
headgroup have been detected with NEXAFS. Distinct
structural differences are observed between the two films.
These changes have been related with their observed mor-
phological differences, leading to a clear relation between
structure and morphology of LB multilayers.
3. Results and discussion
AFM topographic images of both films are shown in
Fig. 1. The CoSt film ð30 ꢂ 30 lm2Þ shows absence of ‘pin-
holes’ (Fig. 1a). Line profile (Fig. 1a inset-I) indicates a
smooth surface except few isolated patches of height
2
˚
ꢁ127 A. A closer scan ð15 ꢂ 15 lm Þ, comparatively void
of these isolated patches, was therefore analyzed (Fig. 1b).
Bearing ratio obtained from AFM analysis (Fig. 1b inset)
shows no decrease of coverage with film thickness except
at the top where there is sharp drop in coverage within a
˚
˚
height of ꢁ30 A (comparable to ML thickness ꢁ27 A). Dis-
tribution of CoSt film height measured from substrate ðz0Þ
showed clear unimodality, with gaussian fit to peak as
2. Experimental details
˚
Nine monolayers (MLs) of CoSt and CdSt were depos-
ited on hydrophilic Si(100) substrate by LB technique
[12]. Fused quartz substrates were used for transmission-
FTIR measurements. Respective metal ions were intro-
duced by adding 0.5 mM solutions of their chloride salts
(CoCl2 ꢀ H2O, Merck; CdCl2 ꢀ H2O, Loba Chemie, 99%) in
a Langmuir trough (KSV instruments, KSV-5000) contain-
ing Milli-Q water (resistivity 18.2 MX cm). pH (ꢁ6.0) was
adjusted by sodium bicarbonate (NaHCO3, Merck, 99%).
Stearic acid (Sigma–Aldrich, 99%) monolayers were spread
from a 0.5 mg/ml chloroform (CCl4, Sigma–Aldrich,
99.9+%) solution. Substrates were made hydrophilic by
keeping them in a solution containing ammonium hydrox-
ide (Merck, 98%), hydrogen peroxide (Merck, 98%) and
Milli-Q water (H2O:NH4OH:H2O2 2:1:1 by volume) for
10–15 min at 100 °C. LB films were deposited at monolayer
pressure of 30 mN/m (measured by Platinum Wilhelmy
plate) at 19 °C with 3 mm/min dipping speed. Drying time
was 10 min and 5 min after first and subsequent strokes
respectively. Films were checked for reproducibility.
It should be mentioned here that although both CdSt
and CoSt LB multilayers could be drawn with ease, CdSt
film coverages were found to vary randomly with every
deposition. CoSt film coverage remained more or less
constant.
127 A, matching exactly with height of isolated patches.
However, for region void of patches, this peak shifts to
˚
ꢁ28 A (Fig. 1a inset-II), consistent with ML thickness,
clearly showing that pinholes present in CoSt films are
strictly confined within topmost layer: an indication of very
good film coverage along z0. Estimated roughness is also
˚
quite low (ꢁ4 A). In sharp contrast to CoSt film, ‘pinhole’
defects are present at CdSt film surface (Fig. 1c and d).
Although defects actually grow in bilayer (BL) steps, line
profile of CdSt (Fig. 1c inset-I) drawn from ð15 ꢂ 15 lm2Þ
image shows V-shaped pinholes due to convolution with
AFM tip. Profile analysis shows presence of pinholes in
each BL of CdSt, of different size and depth due to corre-
sponding cumulative effect (average depth estimated as
˚
248 A). The bearing ratio (Fig. 1d inset) shows a gradual
decrease of coverage-area with z0 - an indication of large
number of pinholes. Moreover noticeable change in slope
˚
of CdSt bearing ratio at ꢁ150 A, indicates enhanced
decrease in film coverage above this height. Thus pinhole
defects, although present from first ML, are greatly
increased in number above the aforesaid height measured
from substrate. This is confirmed from bimodality in z0 dis-
tribution of CdSt (Fig. 1c inset-II). Gaussian fits to distribu-
˚
˚
tion peaks mark their positions as 162 A and 270 A. Higher
˚
z0-value peak (at 270 A) signifies total film thickness
whereas lower peak (at 162 A) corresponds to height above
˚
Surface morphology of films was studied by contact
mode-AFM (Autoprobe CP, Park Scientific). Scans vary-
ing from 15 ꢂ 15 lm2 to 40 ꢂ 40 lm2 were performed in
constant force mode over several regions of film. Low force
constant (1.3 mN) was used to minimize damage. XRR [20]
was done using Versatile X-Ray Diffractometer (VXRD;
which number of pinholes increase greatly, agreeing with
value obtained from bearing ratio. Average surface rough-
˚
ness (including pinholes) is quite high (ꢁ57 A) for CdSt
˚
whereas defect-free regions are much less rough (ꢁ10 A)
showing little out-of plane fluctuation where BLs are pres-
ent. However, it must be borne in mind that the height val-
ues obtained from AFM data include tip convolution effect.
Precise height values were obtained from XRR analysis.
XRR data of the films are shown (open circles) in Fig. 2.
For CoSt (Fig. 2a) 7 Bragg peaks with strong Kiessig
(interference) fringes and for CdSt (Fig. 2b) 11 Bragg peaks
with less prominent Kiessig fringes are obtained, showing
˚
Bruker AXS) with X-ray wavelength k = 1.54 A (Cu Ka
line). FTIR spectroscopy (Perkin–Elmer) [21] was carried
out in transmission and attenuated total reflectance
(ATR) modes at resolution of 4.0 cmꢃ1. NEXAFS spec-
troscopy [22] at oxygen K-edge (530 eV) was carried out
at normal incidence using synchrotron radiation at BEAR
beamline, ELETTRA synchrotron. The spectra were

82
S. Mukherjee et al. / Chemical Physics Letters 451 (2008) 80–87
Fig. 1. AFM studies of the nine monolayer LB films of metal stearates. (a, c): (30 ꢂ 30) lm2 and (b, d): (15 ꢂ 15) lm2 topographic images; (a, c) (insets-I):
corresponding line profiles and (insets-II): Height (z0) distribution data (circles) with corresponding Gaussian fits (line); (b, d) (insets): corresponding
bearing ratios. (a, b) correspond to CoSt and (c, d) correspond to CdSt films.
CdSt to have lower coherence between air/film and film/
substrate interfaces compared to CoSt. This is most prob-
ably due to presence of pinhole defects. Large number of
Bragg peaks indicate both LB multilayers have good out-
of-plane crystalline growth (comparatively better for
CdSt). This apparently indicates no direct correlation
between pinhole defects and out-of-plane crystallinity and
suggests role of in-plane forces in defect formation. XRR
data was analyzed using Parratt formalism [23], where each
film is divided into layers of fixed thickness (d), average
electron density (q) and interfacial roughness (r), and used
as fit parameters [24]. Solid lines in Fig. 2a and b are best-
fit curves. Electron density profiles (EDPs) for the films (q)
as a function of film thickness z, where z ¼ 0 is air),
obtained from fits, are shown in respective insets. EDP of
CdSt show gradual decrease in q with increase in BL num-
ber, from very first monolayer contacting substrate (H, T
and G denote head, tail and gap between two tails respec-
tively). Pinholes reduce film coverage, giving rise to such
˚
EDP, consistent with AFM data. Air/film r-value is 6 A.
EDP of CoSt show no such decrease of coverage to about
3 12 BLs from substrate, and very small decrease in q for
topmost BL. This indicates presence of few pinholes in
CoSt primarily confined to top BL. Air/film r-value is
˚
6 A for CoSt, consistent with AFM results. Thus, for CoSt
we have more compact film growth. However, height of
pinholes in CoSt film obtained from its EDP did not
exactly match the AFM results. This is not surprising, as

S. Mukherjee et al. / Chemical Physics Letters 451 (2008) 80–87
83
100
10-3
10-6
a
CoSt
0.6
c
2
0.3
CoSt
CdSt
0.0
0
100
Z (Å)
200
1
0
0
100
200
300
zBL (Å)
0.0
0.3
0.6
)
0.9
−1
qZ( Å
100
10-3
10-6
b
CdSt
0.8
H
SiO2
G
H
d
120
80
40
0
CdSt
CoSt
H
T
0.4
0.0
T
T
H
T
T
T
T
T
G
G
G
0
100
Z (Å)
200
0
60
120 180 240
zBL (Å)
0.0
0.4
0.8
−1
1.2
qZ( Å
)
Fig. 2. (a, b): X-ray reflectivity data (open circles) with corresponding fits (solid lines), of the deposited LB films (insets: corresponding EDPs (H, T, G
denote head, tail and gap respectively)). (a, b) correspond to CoSt and CdSt films respectively. (c) Calculated head-tail coverage ratio (N0H : NT0 ) versus
height (z0) of the nine monolayer films. (d) Calculated percentage coverage versus height (z0) of the LB films.
size of these pinholes are much smaller than those in CdSt
film hindering the finite width of AFM tip to sense proper
height variation.
headgroup (tail) thickness. Each N - value was normalized
with corresponding bulk value Nb to account for porosity
(pinholes). To eliminate decrease in q due to pinholes,
the ratio N0H:ðNT0 Þ was obtained for each BL, where NH0
ðN0TÞ is normalized value of N for head (tail) and plotted
as a function of BL height ðzBLÞ for both films (Fig. 2c).
Slope v of these curves is a measure of good out-of-plane
To quantify goodness of multilayer growth, assuming
2
˚
uniform molecular cross-sectional area A of 20 A , we cal-
culated number of electrons NH ðNTÞ in Head (Tail) for
each BL as qH ꢂ A ꢂ dH ðqT ꢂ A ꢂ dTÞ, where dH ðdTÞ is
a
b
c
CoSt
(I)
2956
2955
(I)
(II)
1540
881
1115
1700
1473
1463
1472
1464
1411
1175
1179
(II)
CdSt
1411
1397
1311
940
1298
2954
(III)
1702
1689
880
1111
1700
2870
2850
1472
1560
1462
1397
2850
2917
2919
2917
Frequency (cm-1)
2840
2880
2920
2960
3000
Frequency incm-1
1650
1800
720 960 1200
1350 1425 1500 1575
Frequency (cm-1)
Fig. 3. (a) FTIR spectra of bulk stearic acid. Portions (II) and (III) of the spectra have been arbitrarily shifted for clarity. (b) FTIR spectra of (I) CoSt and
(II) CdSt LB films. Y-axis for the selected frequency regions are chosen arbitrarily for better clarity. (c) FTIR spectra of (I) CoSt and (II) CdSt in
transmission mode.

84
S. Mukherjee et al. / Chemical Physics Letters 451 (2008) 80–87
˚
Table 1
reflectivity analysis. BL thickness for CdSt is 50.3 A which
corresponds to normal tail orientation to substrate whereas
Peak assignments of FTIR spectra of the LB films and bulk Stearic Acid
Observed vibrational frequencies (cmꢃ1) for
˚
same for CoSt is 49.6 A, corresponding to 9.6° tilt of tail
from vertical.
Stearic acid
(bulk)
CdSt
CoSt
Assigned to
FTIR-ATR spectroscopy was carried out to explore the
bonding (primarily metal-ion headgroup coordination) and
some aspects of molecular structure of these films (Fig. 3b).
Assigned peaks of these and bulk stearic acid (Fig. 3a) are
tabulated below (Table 1) [25,28]. Absence of OH stretch
around 3400 cmꢃ1 and reduction in C@O stretch near
1700 cmꢃ1 in LB film spectra confirm conversion of free
stearic acid to corresponding metal stearate [25]. To find
metal-ion carboxylate coordination, difference in symmet-
ric ðmsÞ and asymmetric ðmaÞ COO stretching frequencies
ðDÞ are noted [26]. For CoSt ðD ¼ 143Þ the coordination
is bidentate bridging [27], corresponding to one carboxly-
late group per metal ion. However, CdSt has unidentate
coordination ðD ¼ 163Þ with two carboxylate groups per
metal ion.
(3750–3000)
2954 (W)
2917 (S)
2870 (W)
2850 (S)
1702 (VS)
1687 (VS)
–
–
–
mðO–HÞ
2955 (W)
2919 (S)
(VW)
2850 (S)
–
2956 (W)
2917 (S)
(VW)
2850 (S)
–
maðCH3Þ
maðCH2Þ
msðCH3Þ
msðCH2Þ
mðC ¼ OÞ; splitted due
to dimer formation
maðCOOÞ
–
–
1560 (S)
1472 (S)
1540 (S)
1472 (M)
1473 (M)
dðCH2Þ – Scissor; splitted
due
1463 (S)
(1450–1187)a
1431 (M)
1411 (S)
–
1311 (VS)
1298 (VS)
–
1462 (S)
Present
1431 (M)
1404 (M)
1397 (VS)
–
1464 (S)
Present
1430 (M)
1404 (M)
1397 (VS)
–
to crystal field
dðCH2Þ – rock, wag, twist
daðCH3Þ
dða–CH2Þ
dsðCOOÞ
mðC–OHÞ
–
–
mðC–OHÞ
1179 (VS)
1111 (VS)
1175 (VS)
1115 (VS)
mðC–OMꢄÞ
–
mðC–OMꢄÞ
1102 (VW)
940 (S)
–
Convoluted Convoluted mðC–CÞ
Table 2
Analysis of NEXAFS spectra
–
–
dðO–HÞ
880 (S)
881 (S)
dðO–MꢄÞ
Parameters of voigt fit
(m: stretch; d: deformation/bend; a: asymmetric; s: symmetric;
xc
A
wG
wL
Mꢄ ¼ Cd; Co).
CoSt
VS: very strong; S: strong; M: medium; W: weak; VW: very weak.
Refer to text for origin of assignment.
1
2
3
4
532.0
532.7
538.0
542.5
0.20
0.13
0.21
1.30
1.00
1.30
2.50
7.40
0.23
0.50
1.00
0.70
a
No. of band propagations of equally weak intensity.
crystalline growth, v ¼ 0 denoting perfect multilayer stack-
ing irrespective of presence of pinholes. v-values obtained
were ꢁ10ꢃ4 for both, showing excellent crystalline growth.
Percentage film coverage, calculated from tails ðNT ꢂ 100Þ,
versus zBL (Fig. 2d) show linear variation for CoSt but qua-
dratic decay for CdSt, consistent with AFM results. Slope
ðv0 ¼ 0:07Þ of the linear fit indicates compact in-plane
defect free coherent growth in CoSt (v0 ꢁ 0 corresponding
to perfectly coherent growth). Orientation of hydrocarbon
tails were calculated from BL thicknesses obtained from
CdSt
1
2
3
4
5
6
7
531.9
532.7
533.7
535.2
537.8
540.8
543.6
0.22
0.03
0.10
0.02
0.37
0.35
1.65
0.80
0.40
1.00
0.80
2.3
0.18
0.30
1.00
0.80
2.2
4.0
5.5
1.0
2.00
xc: Center; A: amplitude; wG: Gaussian width; wL: Lorenzian width.
xc, wG and wL are in eV, A is in a.u.
a
b
1
π∗
π∗
σ∗
σ∗
6
2
1
4
5
7
3
4
2
3
530
535
540
545
550
530
535
540
545
550
Energy in eV
Energy in eV
Fig. 4. O K-edge NEXAFS spectra of 9 ML (a) CoSt and (b) CdSt LB films. Experimental data (circles) have been fitted with voigt function (dotted lines).
Solid lines are the composite fits.

S. Mukherjee et al. / Chemical Physics Letters 451 (2008) 80–87
85
LB films of fatty acids are known to have orthorhombic
packing with all-trans conformation of acyl chains [1].
Presence of two peaks near 1462 cmꢃ1 and 1474 cmꢃ1 indi-
cate orthorhombic packing [28]. However, intensity of
1462 cmꢃ1 peak is reduced in case of CoSt, as compared
to the equally intense doublets in CdSt. This indicates pos-
sibility of deviation from true orthorhombic packing in
CoSt. This difference in packing may lead to a shift in
methyl stretching frequencies of hydrocarbon tail. To
observe the methyl and methylene stretching frequencies,
transmission-FTIR spectroscopy was carried out with
9ML films deposited on fused quartz, since Si is not trans-
parent in the region of interest (Fig. 3c). The CH2 asym-
metric stretch (observed at 2917 cmꢃ1) showed a down
shift of 2 cmꢃ1 for CoSt compared to CdSt (observed at
2919 cmꢃ1), which is indicative of an increase in packing
density [29], consistent with higher film coverage observed
for CoSt.
O K-edge NEXAFS spectra of CoSt and CdSt films are
shown in Fig. 4 (solid circles). Both spectra have been fitted
(solid line) with voigt function. Fitting parameters are
shown in Table 2. The low energy sharper feature of the
spectra arise due to pꢄ transitions whereas higher energy
broader part denote rꢄ transitions [22]. In CoSt, two rꢄ
Fig. 5. Cartoon showing LB structure of (a) CoSt and (b) CdSt. h represents the tilt angle of the hydrocarbon chain (h ¼ 9:6ꢅ and 0° for CoSt and CdSt
respectively). The metal ion headgroup coordination is shown separately as (I): Bidentate bridging for CoSt and (II): Unidentate for CdSt (M = Cd or Co).

86
S. Mukherjee et al. / Chemical Physics Letters 451 (2008) 80–87
peaks (3 and 4) and in CdSt three such peaks (5, 6 and 7)
are present. The low energy peak near 538 eV (present in
both) is attributed to Si–O bonds coming from native
SiO2 layer of thickness ꢁ25 A [30]. Higher energy peaks
is responsible for deposition of such defect-free LB multi-
layers. Even then, the question as to why these ions attach
to the headgroups in their respective modes is difficult to
answer at this moment. However, it is interesting to note
that atomic number of Co (Z = 27) lies close to number
of electrons present in COO group (22), so that bidentate
structure (in which two metal ions are bridged to two
COO groups) favors uniform electron distribution. Like-
wise, in case of Cd (Z = 48) the unidentate structure (with
one metal ion attached to two COO groups) provides uni-
form electron distribution. From this observation, it can be
proposed that electron distribution near headgroup may
play vital role in deciding headgroup structure. However,
detailed study has to be done in understanding specific
headgroup-metal interactions at air/water interface before
drawing any straightforward conclusion in this respect.
Work is underway along these lines.
˚
at 540.8 eV and 543.6 eV for CdSt correspond to O 1s !
rꢄC–O and O 1s ! rCꢄ @O transitions respectively [31,32].
Presence of both these peaks and also the pꢄ peak (marked
as 1) at 532 eV corresponding to O 1s ! pꢄC@O transition
confirm presence of two asymmetric CO bonds in CdSt
bilayer headgroup. As obtained from FTIR studies, there
is only one Cd ion attached to two carboxylate groups in
the BLs (unidentate coordination). This introduces an
asymmetry in CO bond strength and orientation. The BL
headgroups contain one C@O group along with another
C–O group, latter attached to metal ion. However, due to
delocalization, both bonds have partial pi overlap, but
the two CO bonds of the carboxylate group are not equiv-
alent. On the other hand, single CO sigma peak at 542.5 eV
for CoSt denotes presence of symmetric CO bonds in CoSt
headgroups. Energy of these CO bonds lie midway between
that of C–O and C@O bonds in CdSt. Moreover, similar pꢄ
peak near 532 eV is present. Thus two CO bonds of the car-
boxylate group present in CoSt bilayers are identical. These
are consistent with bidentate bridging mode of metal ion
coordination as obtained from FTIR study, as in this case
symmetric CO bonds have nature intermediate between C–
O and C@O bonds. It is also noted that p peaks in CdSt
have smaller line widths compared to CoSt (Table 2), indic-
ative of lower degree of delocalization consistent with
asymmetric nature of bonds. Low intensity pꢄ peaks (2, 3
and 4) may be due to transition to Rydberg states and
are not assigned [22]. It is to be noted that absence of
Co–O or Cd–O bonds are due to high Z-value of the metals
since NEXAFS spectroscopy is applicable for detection of
species with low Z values (Z < 10) [22].
Acknowledgements
Authors wish to thank Prof. Satyajit Hazra for avail-
ability of VXRD facility in caring out XRR measurements.
We gratefully acknowledge Prof. Debabrata Ghose and
Mr. Puneet Mishra for their time and support in carrying
out AFM measurements.
References
[1] M.C. Petty, Langmuir–Blodgett Films: An Introduction, Cambridge
University Press, Cambridge, 1996.
[2] A. Ulman, Introduction to Ultrathin Organic Films, Academic, New
York, 1991.
[3] G. Roberts, Langmuir–Blodgett Films, Plenum, New York, 1990.
[4] P. Facci, A. Diaspro, R. Ronaldi, Thin Solid Films 327 (1998) 532.
[5] K. Tollner, R. Popovitz-Biro, M. Lahav, D. Milstein, Science 278
(1997) 2100.
[6] C.P. Collier, R.J. Saykally, J.J. Shiang, S.E. Henrichs, J.R. Health,
Science 277 (1997) 1978.
4. Conclusion and outlook
[7] J. Als-Neilsen, D. Jacquemain, K. Kjaer, F. Leveiller, M. Lahav, L.
Leiserowitz, Phys. Rep. 246 (1994) 251.
[8] J.D. Swalen, Thin Solid Films 160 (1988) 197.
[9] J.H. Fendler, Membrane Mimetic Chemistry, Wiley-Interscience
Press, New York, 1982.
AFM and XRR results indicate large reduction of in-
plane defects in CoSt multilayers compared to CdSt, along
with excellent out-of-plane crystalline growth in both. CoSt
film shows bidentate bridging metal ion-carboxylate coor-
dination, as opposed to unidentate coordination in CdSt.
Also CoSt films have higher hydrocarbon tail packing den-
sity; tails tilted at 9.6° to normal (Fig. 5). FTIR and
NEXAFS studies clearly indicate underlying importance
of headgroup structure in defining morphology of LB films.
We propose that the bidentate bridging coordination plays
a key role in stacking of defect-free multilayers. Thus
structure and hence morphology of films is governed by
metal-ion headgroup interaction. Cobalt and Cadmium
show different behavior at air/water interface and have
accordingly been classified as type I and type II respectively
[17]. Moreover they form different 3D structures after col-
lapse [18]. Our results do not disagree with the obtained
difference in trend between these two divalent metal ions.
Perhaps it can be said that it is this inherent difference that
[10] G.G. Roberts, Adv. Phys. 34 (1984) 475.
[11] V. Nitz et al., Phys. Rev. B 54 (1996) 5038.
[12] D.K. Schwartz, Surf. Sci. Rep. 27 (1997) 241.
[13] Th. Geue, M. Schultz, U. Englisch, R. Stommer, U. Pietsch, K.
Meine, D. Vollhardt, J. Chem. Phys. 110 (1999) 8104.
[14] D.Y. Takamoto, E. Aydil, J.A. Zasadzinski, A.T. Ivanova, D.K.
Schwartz, T. Yang, P.S. Cremer, Science 293 (2001) 1292.
[15] S. Kundu, A. Datta, S. Hazra, Chem. Phys. Lett. 405 (2005) 282.
[16] A. Datta, J. Kmetko, C.J. Yu, A.G. Richter, K.S. Chung, J.M. Bai, P.
Dutta, J. Phys. Chem. B 104 (2000) 5797.
[17] J. Kmetko, A. Datta, G. Evmenenko, P. Dutta, J. Phys. Chem. B 105
(2001) 10818.
[18] S. Kundu, A. Datta, S. Hazra, Langmuir 21 (2005) 5894.
[19] M. Sachan, N.D. Walrath, S.A. Majetich, K. Krycka, Chi-Chang
Kao, J. Appl. Phys. 99 (2006) 08C302.
[20] M.K. Sanyal, A. Datta, S. Hazra, Pure Appl. Chem. 74 (2002)
1553.
[21] Brian C. Smith, Fundamentals of Fourier Transform Infrared
Spectroscopy, CRC Press, New York, 1996.

S. Mukherjee et al. / Chemical Physics Letters 451 (2008) 80–87
87
[22] J. Stohr, NEXAFS Spectroscopy, Springer-Verlag, New York, 1996.
[23] L.G. Parratt, Phys. Rev. 95 (1954) 359.
[28] J.F. Rabolt, F.C. Burns, N.E. Schlotter, J.D. Swalen, J. Chem. Phys.
78 (1983) 946.
[24] J.K. Basu, M.K. Sanyal, Phys. Rep. 363 (2002) 1.
[25] William Kemp, Organic Spectroscopy, Palgrave, Hampshire, 2002.
[26] K. Nakamoto, Infrared and Raman Spectra of Inorganic and
Coordination Compounds, Wiley and Sons, New York, 1970.
[27] E.G. Palacios, G. Juarez-Lopez, A.J. Monhemius, Hydrometallurgy
72 (2004) 139.
[29] Y. Lu, J.D. Miller, J. Colloid Interface Sci. 256 (2002) 41.
[30] Chang Ok Kim et al., Langmuir 19 (2003) 4504.
[31] S.G. Urquhart, A. P Hitchcock, A.P. Smith, H. Ade, E.G. Rightor, J.
Phys. Chem. B 101 (1997) 2267.
[32] S.G. Urquhart, H. Ade, J. Phys. Chem. B 106 (2002) 8531.