JOURNAL OF RAMAN SPECTROSCOPY J. Raman Spectrosc. 2007; 38: 323–331 Published online 22 November 2006 in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/jrs.1647
IR, Raman and SERS spectra of 3,5-dinitrosalicylic acid Hema Tresa Varghese,1 C. Yohannan Panicker,2∗ Daizy Philip,3 Joydeep Chowdhury4 and Manash Ghosh5 1 2 3 4 5
Department of Physics, Fatima Mata National College, Kollam, Kerala, 691 001, India Department of Physics, TKM College of Arts and Science, Kollam, Kerala, 691 005, India Department of Physics, Mar Ivanios College, Nalanchira, Thiruvananthapuram, Kerala, 695 015, India Department of Physics, Sammilani Mahavidyalaya, Baghajatin Station, EM Bypass, Kolkata, 700 075, India Department of Spectroscopy, Indian Association for the Cultivation of Science, Jadavpur, Kolkata, 700 032, India
Received 27 June 2006; Accepted 29 August 2006
IR, Raman and surface enhanced Raman scattering (SERS) spectra of 3,5-dinitrosalicylic acid (DNSA) were recorded and analysed. The vibrational wavenumbers were computed by the ab initio method using RHF/6–21G∗ basis and they were found to be in good agreement with the experimental values. The effect of the concentration dependence on the SERS intensity of the molecule was studied. The molecular plane assumes a tilted orientation with respect to the silver surface. Copyright 2006 John Wiley & Sons, Ltd.
KEYWORDS: 3,5-dinitrosalicylic acid; SERS; vibrational spectra; ab initio calculation
INTRODUCTION 3,5-Dinitrosalicylic acid (DNSA) is an unusual example of a proton donor having both carboxyl and hydroxyl groups. DNSA provides one of the best chemical synthons for the construction of hydrogen-bonded structural motifs.1 The acid has provided examples of polymorphism in which associations with solvent molecules such as water, dioxane and tert-butyl alcohol give a variety of hydrogen-bonded molecular assemblies.2,3 The interaction of DNSA with Lewis bases has been extensively studied by Issa et al.,4 – 7 with more than 100 charge-transfer and proton-transfer compounds having been synthesized and characterized by IR spectroscopy. Several of these compounds have unusual physical properties with potential commercial applications, e.g. the electrical properties of the diaminonaphthalene charge-transfer complex.4 – 7 He et al.8 reported the synthesis, crystal structure and magnetic properties of 1D cobalt(II) complex with 1,10-phenanthroline and DNSA ligands using X-ray analysis, IR spectroscopy, X-ray photoelectron spectroscopy (XPS), thermogravimetry-differential thermal analysis (TG-DTA) and elemental analysis. The structural and spectroscopic studies on DNSA complexes of urotropine and dicyclohexylamine were reported by Ng et al.,9 and they have carried out the theoretical analysis of the structure of the complex anion. Structural confirmation of the 3,5-dinitrosalicylate anion coordination ability to metal ions have been reported Ł Correspondence to: C. Yohannan Panicker, Department of Physics, TKM College of Arts and Science, Kollam, Kerala, 691 005, India. E-mail:
[email protected]
Copyright 2006 John Wiley & Sons, Ltd.
by Valigura et al.,10,11 and they have reported the characterization of the complex by elemental micro-analysis, electronic, IR and EPR spectra. DNSA molecules were found in these structures.12 – 14 Nitro derivatives of salicylic acids as a donor in solid state coordination compounds have been studied only sporadically,15 and the study of metal complexes with DNSA in solid state is also rare.16 – 18 The authors have reported the IR, Raman and surface enhanced Raman scattering (SERS) spectra of 4-aminosalicylic acid,19 5-sulphosalicylic acid,20 sodium salicylate21 and methyl salicylate.22 However, there is no report on the IR, Raman and SERS spectra of the title compound. In the present study, the IR, Raman and concentration-dependent SERS studies of DNSA were investigated to get an idea regarding the orientation of the molecule on the silver surface. Theoretical calculations of the vibrational wavenumbers were made using the Gaussian03 software package.23
EXPERIMENTAL DNSA was procured from Sigma-Aldrich, USA. The Fouriertransform infrared (FT-IR) spectrum (Fig. 1) was recorded using a Bruker IFS 66V FT-IR spectrometer with the sample as KBr and polyethylene pellets. Raman spectra (Figs 2–4) were recorded by a Spex model 1403 double monochromator fitted with a holographic grating of 1800 grooves mm1 and a cooled model R928/115 photomultiplier tube (Hamamatsu Photonics, Japan). The samples were placed in a quartz cell and excited with the 514.5 nm radiation from a SpectraPhysics Model 2020–05 argon ion laser at a power of 0.1 W. The operation of the photon counter, data acquisition
324
H. T. Varghese et al.
and analysis were controlled by Spex Datamate 1B. The acquisition time by the spectral element was 0.5 s. The scattered light was focused on to the entrance slit of width 4 cm1 . All the Raman spectra shown here are original raw data directly transferred from the instrument and processed using the Microcal origin version 4.10 16-bit program. They are presented even without single smoothing. A stable silver sol was prepared by the process of Creighton et al.,24 and stored at 5 ° C. All the required solutions were prepared with distilled, deionized water obtained from a Milli-Q-plus system (Millipore).
COMPUTATIONAL DETAILS The vibrational wavenumbers were calculated using the Gaussian03 software.23 The wavenumber values computed at the Hartree-Fock level contain known systematic errors because of not taking into account the electron correlation, and hence the calculated values are multiplied by scaling factors.25 Molecular geometries were fully optimized by Berny’s optimization algorithm using redundant internal coordinates. All optimized structures (Table 1) were confirmed to be minimum energy conformations. Harmonic vibrational wavenumbers were calculated using analytic Figure 3. SERS spectra of DNSA in silver colloid at concentrations (a) 103 M (b) 104 M (c) 105 M and (d) 106 for a freshly prepared colloid–molecule mixture.
M
Figure 1. IR spectrum of DNSA.
Figure 4. SERS spectra of DNSA in silver colloid at concentrations (a) 104 M (b) 105 M and (c) 106 M recorded after 72 h from the preparation of the colloid–molecule mixture. Figure 2. Raman spectrum of DNSA.
Copyright 2006 John Wiley & Sons, Ltd.
J. Raman Spectrosc. 2007; 38: 323–331 DOI: 10.1002/jrs
IR, Raman and SERS spectra of 3,5-dinitrosalicylic acid
second derivatives to confirm the convergence to minimum on the potential surface. At the optimized structure of the examined species (Fig. 5), no imaginary wavenumber modes were obtained, proving that a true minimum on the potential surface was found. The optimum geometry was determined by minimizing the energy with respect to all geometrical parameters without imposing molecular symmetry constraints. The assignment of the calculated wavenumbers is aided by the animation option of the MOLEKEL program, which gives a visual presentation of the vibrational modes.26,27
RESULTS AND DISCUSSION The observed Raman and IR bands with their relative intensities, calculated values and assignments are given in Table 2.
IR and Raman spectra The most characteristic bands in the spectra of nitro compounds are due to NO2 stretching vibrations, which are the two most useful group wavenumbers, not only because of their spectral position but also for their strong intensity.28 In nitro compounds, the antisymmetric NO2 stretching vibrations28 are located in the region 1580 š 80 cm1 . The symmetric NO2 stretching vibrations28 are expected in the region 1380 š 20 cm1 . In substituent nitrobenzenes, s NO2 appears strongly at 1345 š 30 cm1 , in 3-nitropyridine29 at 1350 š 20 cm1 and in conjugated nitroalkenes30 at 1345 š 15 cm1 . Ng et al.9 reported as NO2 at 1535 and s NO2 at 1344 cm1 for DNSA. For DNSA complexes9 s NO2 are reported at 1350 and 1326 cm1 . Smith et al.2 reported as NO2 at 1540 and 1530 cm1 for DNSA. In the present case, the bands observed at 1533 cm1 in the IR
Table 1. Optimized geometrical parameters of 3,5-dinitrosalicylic acid: the atom labelling is according to Fig. 5 ˚ Bond lengths (A) C1 –C2 C2 –C3 C3 –O9 C5 –C6 C8 –O18 O9 –H19 N12 –O13 O18 –H19 Bond angles(° ) A(2,1,6) A(1,2,3) A(2,3,4) A(3,4,5) A(4,5,6) A(1,6,5) A(2,8,18) A(3,9,17) A(15,10,16) A(13,12,14) Dihedral angles(° ) D(6,1,2,3) D(7,1,2,8) D(7,1,6,5) D(1,2,3,9) D(1,2,8,18) D(3,2,8,20) D(9,3,4,5) D(4,3,9,17) D(10,4,5,6) D(3,4,10,16) D(4,5,6,1) D(11,5,6,12) D(5,6,12,13) D(20,8,18,19)
1.3811 1.4056 1.3245 1.3719 1.3201 1.8012 1.195 0.9649 120.7 119.5 118.2 121.9 118.4 121.3 118.2 107.9 124.4 126.8 0.0 0.0 180.0 180.0 180.0 180.0 180.0 0.0 180.0 180.0 0.0 0.0 0.0 180.0
Copyright 2006 John Wiley & Sons, Ltd.
C1 –C6 C2 –C8 C4 –C5 C5 –H11 C8 –O20 N10 –O15 N12 –O14 – A(2,1,7) A(1,2,8) A(2,3,9) A(3,4,10) A(4,5,11) A(1,6,12) A(2,8,20) A(4,10,15) A(6,12,13) A(8,18,19) D(6,1,2,8) D(2,1,6,5) D(7,1,6,12) D(8,2,3,4) D(1,2,8,20) D(2,3,4,5) D(9,3,4,10) D(3,4,5,6) D(10,4,5,11) D(5,4,10,15) D(4,5,6,12) D(1,6,12,13) D(5,6,12,14) –
1.3853 1.5178 1.3843 1.0707 1.181 1.2158 1.1927 – 118.8 115.5 119.0 120.8 120.6 119.7 119.5 117.7 116.6 111.5 180.0 0.0 0.0 180.0 0.0 0.0 0.0 0.0 0.0 180.0 180.0 180.0 180.0 –
C1 –H7 C3 –C4 C4 –N10 C6 –N12 O9 –H17 N10 –O16 O1 –H17 – A(6,1,7) A(3,2,8) A(4,3,9) A(5,4,10) A(6,5,11) A(5,6,12) A(18,8,20) A(4,10,16) A(6,12,14) – D(7,1,2,3) D(2,1,6,12) D(1,2,3,4) D(8,2,3,9) D(3,2,8,18) D(2,3,4,10) D(2,3,9,17) D(3,4,5,11) D(3,4,10,15) D(5,4,10,16) D(11,5,6,1) D(1,6,12,14) D(2,8,18,19) –
1.072 1.4062 1.4556 1.4596 0.9828 1.183 1.6632 – 120.5 125.1 122.7 117.3 121.0 119.0 122.3 117.9 116.6 – 180.0 180.0 0.0 0.0 0.0 180.0 180.0 180.0 0.0 0.0 180.0 0.0 0.0
J. Raman Spectrosc. 2007; 38: 323–331 DOI: 10.1002/jrs
325
326
H. T. Varghese et al.
Table 2. (Continued)
Figure 5. Optimized geometry of 3,5-dinitrosalicylic acid.
Table 2. Wavenumbers (cm1 ) and band assignments Calculated (cm1 )
IR (cm1 )
Raman (cm1 )
SERS (103 M) (cm1 )
Assignments
3517 3213 3095 3083 1696 1592 1573 1537 1533 1413 1382 1365 1351 1326 1319 1265 1227 1163 1155 1137 1069 1029 997 931 893 812 801 785
3566 m 3420 s 3446 m 3236 s 3106 m – – 3085 s 1674 s – 1607 s 1592 m – – – 1538 w 1533 s 1534 w 1454 s 1439 w 1375 m – – – – 1348 vs 1338 s 1334 s 1306 w – 1252 s 1289 m – – 1177 s 1179 w 1156 sh 1155 w – – 1088 s 1086 w – – – – 939 m 936 w 915 s – 835 m – 800 sh 798 m – –
– – – – 1645 s, 1631 s 1602 m 1557 m – – 1487 s – – – 1319 s – 1272 w – 1173 w 1148 vs – – 1024 w – 943 w – – 826 sbr –
OHc OHh CH CH C O Ph Ph as NO2 as NO2 Ph, υOHh C–Oc υOHh s NO2 s NO2 Ph C–Oh υCH Ph Ring breathing υCH υCH CH CH C–N υNO2 υPh υNO2 υO C–OH
Copyright 2006 John Wiley & Sons, Ltd.
Calculated (cm1 )
IR (cm1 )
Raman (cm1 )
775 763 712 706 685 665 654 553 529 526 463 456 410 378 345 332 330 329 210 191 154 129 89 66 59 58
– 743 m 717 s – 696 s – 648 w 550 vw 531 w – 495 m 463 w – – – – – – – – – – – – – –
– 740 w 714 w – 694 w – – – – – – – 407 w – – – – 329 w 210 w – – – – – – –
SERS (103 (cm1 ) – – – – – 668 w – – – – 479 w – – 368 w – – – – – – – – – – – –
M)
Assignments Ph(X) ωNO2 ωNO2 Ph OHh υPh, C O υPh Ph υPh(X) Ph NO2 NO2 υPh(X) COOH υPh(X) υPh(X) Ph(X) υPh(X) Ph(X) tNO2 , tCOOH tNO2 tPh, t NO2 tPh, t NO2 tNO2 tNO2 tNO2
, stretching; υ, in-plane bending; , out-of-plane bending; ω, wagging; , rocking; t, torsional; X, substituent-sensitive; Ph, phenyl; s, strong; w, weak; br, broad; v, very; m, medium; sh, shoulder. Subscript: as, asymmetric; s, symmetric; c, carboxyl group; h, hydroxyl group.
spectrum and 1538 and 1534 cm1 in the Raman spectrum are assigned as as NO2 mode. The s NO2 modes are observed at 1338 cm1 in the IR spectrum and at 1348 and 1334 cm1 in the Raman spectrum. The ab initio calculations give 1537, 1533 and 1351, 1326 cm1 as the antisymmetric and symmetric NO2 modes, respectively. The NO2 scissors28 occur in the range 850 š 60 cm1 when conjugated to C C or aromatic molecules, according to some investigators31 – 33 with a contribution of the CN which is expected near 1120 cm1 . For nitrobenzene, υNO2 is reported28 at 852 cm1 , for H2 C CHNO2 at 890 cm1 and for 1,3-dinitrobenzene at 904 and 834 cm1 . For the title compound the bands at 915 and 800 cm1 in the IR spectrum and at 798 cm1 in the Raman spectrum are assigned to υNO2 modes. The theoretically calculated values are 893 and 801 cm1 .
J. Raman Spectrosc. 2007; 38: 323–331 DOI: 10.1002/jrs
IR, Raman and SERS spectra of 3,5-dinitrosalicylic acid
In aromatic compounds the wagging mode ωNO2 is assigned at 740 š 50 cm1 with a moderate to strong intensity, a region in which CH also is active.28 ωNO2 is reported at 701 and 728 cm1 for 1,2-dinitrobenzene and at 710 and 772 cm1 for 1,4-dinitrobenzene.28 For the title compound, the bands at 743 and 717 cm1 in the IR spectrum and at 740 and 714 cm1 in the Raman spectrum are assigned as ωNO2 modes. The HF calculation gives 763 and 712 cm1 as the ωNO2 modes. For dinitro compounds, the rocking modes28 of NO2 are active in the region 440 š 70 cm1 . These modes are seen at 495 and 463 cm1 in the IR spectrum and at 463 and 456 cm1 theoretically. Carboxylic acids are best characterized by the OH stretch, the C O stretch and the OH out-of-plane deformation. The C O stretching vibration in the spectra of carboxylic acids28 gives rise to a band in the region 1725 š 65 cm1 . In the spectra of salicylic acid, Volovsek et al.34 have reported the C O stretch at 1637 (Raman) and at 1600 cm1 (IR). We have reported C O at 1631 cm1 (Raman) and 1652 cm1 (IR) for sodium salicylate;21 1625 cm1 (Raman) and 1649 cm1 (IR) for 4-aminosalicylic acid;19 1665 cm1 for 5-sulpho salicylic acid;20 and 1683 cm1 for methyl salicylate.22 In the present case, we have observed a strong band at 1674 cm1 in the IR spectrum and 1696 cm1 theoretically. Two bands arising from the C–O stretching and O–H bending appear in the spectra of carboxylic acids near 1320–1210 cm1 and 1440–1395 cm1 , respectively.35 Both of these bands involve some interaction between C–O stretching and in-plane C–O–H bending. The C–Oc mode is reported at 1377 cm1 for sodium salicylate21 and at 1391 cm1 for 4-aminosalicylic acid.19 For the title compound, the band observed at 1375 cm1 (IR) and 1382 cm1 (HF) is assigned as the C–Oc mode. He et al.8 reported the υO–Hh of the phenolic group at 1485 cm1 and C–Oh at 1255 cm1 for 3,5-dinitrosalicylate ligands. In the present case, the band at 1454 cm1 in the IR, 1439 cm1 in the Raman spectrum and 1413 cm1 by HF is assigned as υO–Hh , which is not pure but contains contribution from Ph modes. The band at 1252 cm1 in the IR spectrum, 1289 cm1 in the Raman spectrum and 1265 cm1 by HF is assigned as the C–Oh mode. The aromatic CH stretching vibrations28 absorb weakly to moderately between 3120 and 3000 cm1 . The highest values near 3120 cm1 are observed in the spectra of benzene substituted with CF3 , F, O2 N and C N.28 For the title compound, a medium-intensity band is observed at 3106 cm1 in the IR spectrum and a strong band at 3085 cm1 in the Raman spectrum. The theoretically calculated values are 3095 and 3083 cm1 . The benzene ring possesses six ring stretching vibrations, of which the four with the highest wavenumbers occurring near 1600, 1580, 1490 and 1440 cm1 are good group vibrations.28 With heavy substituents, the bands tend to shift to somewhat lower wavenumbers, and the greater
Copyright 2006 John Wiley & Sons, Ltd.
the number of substituents on the ring, the broader the absorption regions.28 In the case of C O substitution, the band near 1490 cm1 can be very weak.28 The fifth ring stretching vibration is active near 1315 š 65 cm1 , a region that overlaps strongly with that of the CH in-plane deformation.28 The sixth ring stretching vibration, the ring breathing mode, appears as a weak band near 1000 cm1 in mono-, 1,3-di- and 1,3,5-tri-substituted benzenes. In the otherwise substituted benzenes, however, this vibration is substituent-sensitive and difficult to distinguish from other modes. For the title compound, the Ph modes are observed at 1607, 1454, 1306, 1177 and 1156 cm1 in the IR spectrum and at 1592, 1439, 1179 and 1155 cm1 in the Raman spectrum. The HF calculation showed this mode at 1592, 1573, 1413, 1319, 1163 and 1155 cm1 . The out-of-plane and in-plane deformations28 of the phenyl ring are observed below 1000 cm1 and these modes are not pure but contain a significant contribution from other modes and are substituent-sensitive. The aromatic ring in DNSA is somewhat irregular and the spread of bond distance ˚ which is similar to the spread reported by is 1.3719–1.4062 A, 2 Smith et al. The carbon–oxygen distances unambiguously define the single and double bonds in the carboxylate ˚ and group (C8–O18 D 1.3201 and C8–O20 D 1.181 A) are in agreement with the values given by Ng et al.9 The ˚ is carbon–oxygen (phenoalate) C3–O9 distance (1.3245 A) ˚ in agreement with that (1.3295 A) reported by Smith et al.,2 ˚ found among and is less than the average distance of 1.362 A 36 phenols.
SERS spectra Of the spectra at different concentrations, the SERS spectrum at 103 M (Fig. 3(a)) is the prominent one. The prominent bands are observed at 1645, 1631, 1602, 1557, 1487 and 1319 cm1 . The relative intensities from SERS spectra are expected to differ significantly from the normal Raman spectrum owing to specific surface selection rules.37 The surface selection rule suggests that for a molecule adsorbed flat on the silver surface, its out-of-plane vibrational modes will be more enhanced when compared with its in-plane vibrational modes and vice versa when it is adsorbed perpendicular to the surface.37,38 It is further seen that vibrations involving atoms that are close to the silver surface will be enhanced. When the wavenumber difference between Raman bands in the normal and SERS spectra is not more than 5 cm1 , the molecular plane will be perpendicular to the silver surface.39 In the SERS spectrum of 2-amino,5 nitropyridine,40 the symmetric NO2 stretching mode corresponds to the most intense band, which appears broad and significantly downshifted from 1344 to 1326 cm1 , suggesting a binding to silver through the lone pairs of the oxygen atom. Carrosco et al.41 observed the a NO2 band in the SERS spectrum at ¾1500 cm1 with medium intensity, which demonstrates the importance of the nitro group in regard to the interaction
J. Raman Spectrosc. 2007; 38: 323–331 DOI: 10.1002/jrs
327
328
H. T. Varghese et al.
with the metal. Further, they observed the enhancement of Ph modes, revealing that the molecule is oriented perpendicular to the metal surface, whereas the changes that occur in the nitro group indicate that the interaction occurs through O atoms of the nitro moiety. The interaction induces a -electronic redistribution primarily around both the nitro group and the aromatic portion in the vicinity of the substitution site. Also, Gao and Weaver42 observed broadening and a downshift of the corresponding band of nitrobenzene adsorbed on gold via a nitro group. For the title compound, the symmetric stretching mode of NO2 seen at 1334 cm1 in the normal Raman spectrum appears at 1319 cm1 in the SERS spectrum. A charge transfer from the oxygen atoms of the NO2 group to the metal is evidenced by the marked downshift of the symmetric stretching of the NO2 group as detected by the SERS.40 – 42 Interaction through the NO2 group was also supported by the presence of modes at 943, 826 and 479 cm1 . According to the surface selection rule, vibrations involving atoms that are close to the silver surface will be enhanced.37,38 In the SERS spectrum of the title compound, the aromatic CH stretching vibrations are absent, which suggests that the phenyl ring may be in a position close to being flat on the silver surface.39,43,44 It has also been documented in the literature45 that when a benzene ring moiety interacts directly with a metal surface, the ring breathing mode has to be red-shifted by ¾10 cm1 along with substantial band-broadening in the SERS spectrum. Neither a substantial red shift nor significant band-broadening was identified in the SERS spectrum of the title compound, implying that the probability of a direct ring -orbital to metal interaction should be low, in accordance with a tilted position of the ring. In the case of crotonic acid, there is no band at the wavenumber of the C O stretching band in the normal Raman spectrum and a strong band appears at 1635 cm1 in the SERS spectrum.46 A similar strong band appears in the SERS spectra of saturated carboxylic acids on silver sols.46 For the title compound, the C O stretching band is absent in the normal Raman spectrum but is present at 1645 and 1631 cm1 in the SERS spectrum. Of the adsorbed species, the wavenumber of the C O stretch may be lowered by the interaction between the molecules and the surface. It is also possible that the 1631 cm1 band arises from the C C stretch or includes a contribution from this motion.47 Also, the bands at 368 and 668 cm1 in the SERS spectrum support the interaction between the carboxylate group and the silver surface. The Ph stretching vibrations are present in the SERS spectrum at 1602, 1557, 1487 and 1173 cm1 . In the SERS spectrum, substantial band-broadening is observed for the benzene ring vibrations as in 3-aminosalicylic acid.47 This indicates a somewhat flat orientation on the silver surface.47 The out-of-plane CH vibration observed at 1024 cm1 in the SERS spectrum, which is absent in the normal Raman spectrum, indicates a surface -interaction in accordance
Copyright 2006 John Wiley & Sons, Ltd.
with a somewhat flat orientation on the silver surface.19,21 For sodium salicylate21 the C–Oh stretch of the hydroxyl group is present in the SERS spectrum at 1252 cm1 and is red-shifted by 9 cm1 . This is indicative of the nearness of the OH group also to the metal surface. In the present case the C–Oh appears at 1272 cm1 in the SERS spectrum and red-shifted by 17 cm1 . It may happen that C O of the COO–Ag group forms a strong hydrogen bond with the (C)–OH group in the adsorbed state. Keeping in mind the behaviour observed for molecules containing -systems adsorbed flat on metal surfaces,38,42,48,49 a purely parallel orientation of the molecular skeleton with respect to the surface appears unlikely in the light of the following arguments: The aromatic CH stretching vibrations do not show up with any significant intensity in the SERS spectrum. The ring breathing mode in the SERS spectrum at 1148 cm1 shows a small red shift, contrary to the sizeable wavenumber shift reported for analogous modes in systems where a parallel orientation is predominant.38,42 On the other hand, a substantial shift found for other vibrational modes having contributions from the phenyl ring stretching vibrations on going from normal to the SERS spectra is consistent with the interaction of the phenyl ring with the surface, which is expected for a tilted orientation. The above argument is supported by the presence of out-of-plane modes at 1024, 668 and 368 cm1 and in-plane modes at 668 and 398 cm1 in the SERS spectrum. Most of the ring modes are found to be broader in the SERS. The shifts observed for the -ring modes could also be attributed to the interaction of the -electrons with the surface.38,50,51
Concentration dependence of the SERS spectra of DNSA For the SERS study, the solution of DNSA was added to the silver hydrosol in measured quantities to attain the desired small step concentration change of DNSA in the hydrosol. The adsorbate concentration dependence of SERS bands on silver hydrosol, in general, arises from the surface coverage. The SERS signal increases as the concentration of the adsorbate molecule is lowered, attains a maximum at a particular concentration or a particular range of concentrations and then decreases again with a further decrease in concentration.52 It is now well established that on the silver colloidal surface a maximum enhancement of the Raman signal is observed when a monolayer of the adsorbate molecule is formed on the surface, and as multilayers are formed the SERS signal decreases.53,54 This conjecture assumes that the orientation of the molecule does not change with the surface coverage and the predicted behaviour holds for all the SERS bands.52 However, there have been a number of reports in recent years that suggest a change of orientation of the adsorbed molecule with surface coverage and therefore with adsorbate concentration in the sol.55 – 58 The effect of concentration on SERS in relation to the orientation of the molecule with respect to the metal
J. Raman Spectrosc. 2007; 38: 323–331 DOI: 10.1002/jrs
IR, Raman and SERS spectra of 3,5-dinitrosalicylic acid
surface has been reported.57,59,60 In a recent paper on SERS of 1H Indazole on silver sols,59 the authors have noticed changes of relative intensities of bands, particularly the outof-plane modes, as a function of concentration. An increase in relative intensities of the out-of-plane modes has been interpreted to indicate that the molecules assume a more tilted orientation upon lowering the concentration of the adsorbate. A complete change in orientation from end-on to flat has also been reported.57 To investigate the effect of the concentration on the orientation in the present study, the SERS spectra of DNSA at various dilution levels ranging from 103 to 106 M were recorded. The spectra at various concentrations are shown in Fig. 3(a) to (d). No significant changes in the nature of the spectra or line positions were observed except for the lower concentration of 106 M. Maximum enhancement was obtained with the 103 M solution. The C O stretching vibration is present in the SERS spectrum only at 103 M and is absent in other spectra, which indicates the interaction between the carboxylic group and the metal surface at this concentration. At 105 M, a strong band is observed at 1343 cm1 in the SERS spectrum (s NO2 ), which is absent in other SERS spectra and is present in the normal Raman spectrum at 1348 cm1 . The presence of this band and the absence of C D O in the SERS spectrum at 105 M show that at this concentration, the nitro group is close to the metal surface rather than the carboxylic group. For all concentrations, the Ph modes appear in the SERS spectra and are observed with wavenumber shifts >5 cm1 from the corresponding bands in the normal Raman spectrum. As the concentration decreases, the intensity of these bands decreases to a minimum at 105 M, and then increases. The ring breathing mode is seen in the SERS spectrum at 1148 cm1 only for the concentration of 103 M, which is observed as a weak band at 1155 cm1 in the normal Raman spectrum. The Ph band seen at 1439 cm1 in the normal Raman spectrum has contributions from υOHh also. Two bands are observed in the SERS spectrum at 1485 and 1453 cm1 for 104 M and at 1488 and 1449 cm1 for 105 M. The higher wavenumber corresponds to the Ph mode and the lower one corresponds to the υOHh mode. The band at 1024 cm1 in the SERS spectrum at 103 M is the CH mode, which is absent for other concentrations and in the normal Raman spectrum, and the presence of the in-plane modes suggests that at 103 M the molecule has a tilted orientation. As the concentration decreases, the presence of in-plane modes in the SERS spectrum suggests that the orientation changes from the tilted to the flat one.57 The in-plane υCH mode at 1089 cm1 104 M that is absent at 103 M supports this argument. The slight broadening of most of the phenyl stretching vibrations and the presence of the in-plane and out-of-plane modes in the entire range of concentrations suggest a possible slight tilt of the molecule, leading to the
Copyright 2006 John Wiley & Sons, Ltd.
interaction of the -electrons to the ring system with the surface.51 Figure 4(a)–(c) shows the spectra of DNSA recorded after 72 h from the preparation of the colloid–molecule mixture for concentrations ranging from 104 to 106 M. Recently, several authors have reported an unexpected SERS intensity decrease for concentration change of molecules adsorbed on silver colloids, including nucleic acids bases,61,62 aminonaphthalene dyes63 and pefloxacin.64 Sanchez-Cortes et al.62 observed a time evolution of SERS intensity of 1methyl-cytosine and 1,5-dimethylcytosine in silver colloids and they explained this observation by a time-dependent aggregation of the colloidal substrate, which leads to variations in the morphology of the aggregates. It is indeed well known that the crucial parameters influencing the intensities obtained from the metal colloids are the size and morphology of the proper colloidal clusters.65 Since the adsorbates must reduce an electrical repulsive barrier of isolated colloidal particles to induce their aggregation, the latter strongly depends on the nature and concentration of adsorbate and on the colloidal surface potential as well.66,67 The SERS signal of chemisorbed mercaptoethanol68 was found to change as a function of time after addition of mercaptoethanol to the hydrosol and to depend on the concentration. Mercaptoethanol induces the aggregation of the hydrosol in the course of chemisorption. Both processes, chemisorption and aggregation, influence the intensity of the SERS signal. Chemisorption increases the number of molecules contributing to the SERS signal. Aggregation of hydrosol changes the enhancement factor of the SERS-active system. Formation of small and middle-sized aggregates improves the SERS properties of the hydrosol, whereas further aggregation leads to a decrease in enhancement mainly due to fast sedimentation of aggregated silver particles. Besides, considerable aggregation of the hydrosol particle may reduce the overall silver surface area accessible to the molecule and therefore it also decreases the SERS signal. In the present case, as the concentration decreases, the intensity of the bands decreases. The presence of in-plane and out-of-plane modes in the entire range suggests that the orientation is the same as in the case of freshly prepared one. The presence of modes corresponding to the nitro moiety indicates that the interaction between the metal surface and the molecule occurs mainly through the nitro group. The outof-plane CH mode is seen at 1024 cm1 as a weak band that is absent in the fresh spectra for these concentration ranges, which indicates a possible orientation change from the tilted to the flat one. Also, at 105 M, the Ph mode is observed at 554 cm1 , which supports this argument. It can bee seen that the SERS spectra change with time. No significant changes in the nature of the spectra or line positions were observed except for the concentration of 106 M, and the spectra is somewhat similar to the spectra obtained from the freshly prepared colloid–molecule mixture.
J. Raman Spectrosc. 2007; 38: 323–331 DOI: 10.1002/jrs
329
330
H. T. Varghese et al.
CONCLUSIONS The FT-IR and FT-Raman and concentration-dependent SERS spectra of DNSA were studied. The molecular geometry and wavenumbers were calculated using the HartreeFock method with the 6–21GŁ basis set. The observed wavenumbers were found to be in agreement with the calculated values. The SERS signal intensity was found to change as a function of time after the addition of DNSA to the hydrosol and to depend on concentration. For both the freshly prepared and 72-h-aged colloid–molecule mixture, the presence of in-plane and out-of-plane modes in the entire range of concentrations suggests a tilted orientation.
Acknowledgements C.Yohannan Panicker (Teacher fellowship) and Hema Tresa Varghese (minor research grant) would like to thank the University Grants Commission, India.
REFERENCES 1. Smith G, Wermuth UD, White JM. Acta Crystallogr. 2001; E57: 1036. 2. Smith G, Lyncy DE, Byriel KA, Kennard CHL. Aust. J. Chem. 1995; 48: 1133. 3. Kumar VSS, Kuduva SS, Desiraju GR. J. Chem. Soc., Perkin Trans. 1999; 2: 1069. 4. Issa YM, Hindawey AM, Issa RM, Nassar AMG. Rev. Roum. Chim. 1980; 25: 1535. 5. Issa YM, Hindawey AM, El-Kholy AE, Issa RM. Gazz. Chim. Ital. 1981; 111: 27. 6. Hindawey AM, Nassar AMG, Issa RM, Issa YM. Indian J. Chem. 1980; 19A: 615. 7. Ayad MMY, Amer SA, Issa RM. Thermochim. Acta 1989; 140: 277. 8. He X, Li Y-N, Li G-H, Li Y-Z, Zhang P, Xu J-N, Wang Y. Inorg. Chem. Commun. 2005; 8: 983. 9. Ng SW, Naumov P, Drew MGB, Wojciechowski G, Brzezinski B. J. Mol. Struct. 2001; 595: 29. 10. Valigura D, Melnik M, Koman M, Martiska L, Korabik M, Mrozinsky J, Glowiak T. Polyhedron 2004; 23: 2447. 11. Valigura D, Melnik M, Koman M, Martiska L, Korabik M, Mrozinsky J, Glowiak T. Inorg. Chem. Commun. 2004; 7: 548. 12. Smith G, Lynch DE, Byriel KA, Kennard CHL. Aust. J. Chem. 1997; 50: 727. 13. Smith G, Coyone MG, White JM. Aust. J. Chem. 2000; 53: 203. 14. Kumar VSS, Kuduva SS, Desiraju GR. Acta Crystallogr. 2002; 58E: 865. 15. Lajumen LHJ, Kokkonen P, Nissi A, Ruotsalainen H. Thermochim. Acta 1984; 72: 219. 16. Anjaneyulu Y, Rao NVS, Prabhakara Rao R, Murthy LN, Pisipati VGKM. Indian J. Pure Appl. Phys. 1986; 24: 256. 17. Anjaneyulu Y, Rao NVS, Prabhakara Rao R, Pisipati VGKM. Proc. Indian Acad. Sci (Chem. Sci.) 1986; 97: 177. 18. Ranford JD, Sadler PJ, Tocher DA. J. Chem. Soc., Dalton Trans. 1993; 22: 3393. 19. Panicker CY, Varghese HT, Johan A, Philip D, Istvan K, Keresztury G. Spectrochim. Acta 2002; 58A: 281. 20. Varghese HT, Panicker CY, Philip D. J. Raman Spectrosc. 2006; (in press). DOI: 10.1002/jrs.1644. 21. Philip D, John A, Panicker CY, Varghese HT. Spectrochim. Acta 2001; 57A: 1561. 22. Varghese HT, Panicker CY, Philip D, Mannekutla JR, Inamdar SR. Spectrochim. Acta 2006; (in press). DOI: 10.1016/j.saa.2006.08.026.
Copyright 2006 John Wiley & Sons, Ltd.
23. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA, Cheeseman JR, Montgomery JA Jr, Vreven T, Kudin KN, Burant JC, Millam JM, Iyengar SS, Tomasi J, Barone V, Mennucci B, Cossi M, Scalmani G, Rega N, Petersson GA, Nakatsuji H, Hada M, Ehara M, Toyota K, Fukuda R, Hasegawa J, Ishida M, Nakajima T, Honda Y, Kitao O, Nakai H, Klene M, Li X, Knox JE, Hratchian HP, Cross JB, Adamo C, Jaramillo J, Gomperts R, Stratmann RE, Yazyev O, Austin AJ, Cammi R, Pomelli C, Ochterski JW, Ayala PY, Morokuma K, Voth GA, Salvador P, Dannenberg JJ, Zakrzewski VG, Dapprich S, Daniels AD, Strain MC, Farkas O, Malick DK, Rabuck AD, Raghavachari K, Foresman JB, Ortiz JV, Cui Q, Baboul AG, Clifford S, Cioslowski J, Stefanov BB, Liu G, Liashenko A, Piskorz P, Komaromi I, Martin RL, Fox DJ, Keith T, Al-Laham MA, Peng CY, Nanayakkara A, Challacombe M, Gill PMW, Johnson B, Chen W, Wong MW, Gonzalez C, Pople JA. Gaussian03, Revision C.02. Gaussian Inc.: Wallingford, CT, 2004. 24. Creighton JA, Blatchford CG, Albrecht MG. J. Chem. Soc., Faraday Trans. 1979; 75(2): 790. 25. Foresman JB, Frisch E. Exploring chemistry with electronic structure methods; a guide to using gaussian. Gaussian, Inc: Pittsburg, PA, 1996. 26. Flukiger P, Luthi HP, Portmann S, Weber J. MOLEKEL 4.3. Swiss Centre for Scientific Computing: Manno, 2000–2002. 27. Portman S, Luthi HP. Chimia 2000; 54: 766. 28. Roeges NPG. A Guide to the Complete Interpretation of Infrared Spectra of Organic Structures. Wiley: New York, 1994. 29. Perjessy A, Rasala D, Tomasik P, Gawinecki R. Collect. Czech. Chem. Commun. 1985; 50: 244. 30. Brown JF Jr. J. Am. Chem. Soc. 1955; 77: 6341. 31. Green JHS, Kynanston W, Lindsey AS. Spectrochim. Acta 1961; 17: 486. 32. Exner O, Kovac S, Solcaniova E. Collect. Czech. Chem. Commun. 1972; 37: 2156. 33. Ranga Rao DS, Thyagarajan G. Indian J. Pure Appl. Phys. 1978; 16: 941. 34. Volovsek V, Colombo L, Furri K. J. Raman Spectrosc. 1983; 4: 347. 35. Silverstein RM, Webster FX. Spectrometric Identification of Organic Compounds, (6th edn). John Wiley: Asia, 2003. 36. Allen FH, Kennard O, Watson DG, Brammer L, Orpen AG, Taylor R. J. Chem. Soc., Perkin Trans. 2 1987; 12: S1. 37. Creighton JA. Spectroscopy of Surfaces – Advances in Spectroscopy, vol. 16. Wiley: New York, 1988; 37, chapt 2. 38. Gao X, Davies JP, Weaver MJ. J. Phys. Chem. 1990; 94: 6858. 39. Levi G, Patigny J, Massault JP, Aubard J. In 13 th International Conference on Raman Spectroscopy, Wurzberg, 1992; 652. 40. Maurizio MM. Vib. Spectrosc. 2002; 29: 229. 41. Carrasco EA, Compos-Vallette FM, Leyton P, Diaz G, Clavijo RE, Garcia-Ramos JV, Inostroza N, Domingo C, SanchezCortes S, Koch R. J. Phys. Chem. 2003; 107: 9611. 42. Gao P, Weaver MJ. J. Phys. Chem. 1985; 89: 5040. 43. Creighton JA. Surf. Sci. 1983; 124: 209. 44. Moskovits M, Suh JS. J. Am. Chem. Soc. 1986; 108: 4711. 45. Gao P, Weaver KJ. J. Phys. Chem. 1989; 93: 6205. 46. Sun K, Wan C, Xu G. J. Raman Spectrosc. 1989; 20: 267. 47. Nogueira HIS. Spectrochim. Acta 1998; 54A: 1461. 48. Patterson ML, Weaver MJ. J. Phys. Chem. 1985; 89: 5046. 49. Kwon YJ, Son DH, Ahn SJ, Kim MS, Kim K. J. Phys. Chem. 1994; 98: 8481. 50. Cho SH, Han HS, Jang DJ, Kim K, Kim MS. J. Phys. Chem. 1995; 99: 10594. 51. Thomas S, Venkateswaran S, Kapoor S, Cunha RD, Mukherjee T. Spectrochim. Acta 2004; 60A: 25. 52. Chowdhury J, Ghosh M. J. Raman Spectrosc. 2004; 35: 1023. 53. Chowdhury J, Ghosh M, Misra TN. Spectrochim. Acta 2000; 56A: 2107.
J. Raman Spectrosc. 2007; 38: 323–331 DOI: 10.1002/jrs
IR, Raman and SERS spectra of 3,5-dinitrosalicylic acid
54. Chowdhury J, Ghosh M, Misra TN. J. Colloid Interface Sci. 2000; 228: 372. 55. Chowdhury J, Ghosh M, Pal P, Misra TN. J. Colloid Interface Sci. 2003; 263: 318. 56. Kim SK, Joo TH, Suh SW, Kim MS. J. Raman Spectrosc. 1986; 17: 381. 57. Moskovits M, Suh JS. J. Phys. Chem. 1988; 92: 6327. 58. Joo TH, Kim K, Kim MS. J. Phys. Chem. 1986; 90: 5816. 59. Pergolese B, Bigotto A. Spectrochim. Acta 2001; 57A: 1191. 60. Chen W, Xu Q, Jiang H, Xu Z. Spectrosc. Lett. 2000; 33: 69. 61. Wentrup-Byrne E, Sarinas S, Fredericks PM. Appl. Spectrosc. 1993; 47: 1192.
Copyright 2006 John Wiley & Sons, Ltd.
62. Sanchez-Cortes S, Garcia-Ramos JV, Morcillo G. J. Colloid Interface Sci. 1994; 167: 428. 63. Jones JC, McLaughlin C, Littlejohn D, Sadler DA, Graham D, Smith WE. Anal. Chem. 1999; 71: 596. 64. Lecomte S, Moreau NJ, Manfait M, Aubard J, Baron MH. Biospectroscopy 1995; 1: 423. 65. Shalaev VM. Phys. Rep. 1996; 272: 61. 66. Weitz DA, Lin MY, Sandrolf CJ. Surf. Sci. 1985; 158: 147. 67. Feilchenfeld H, Siiman O. J. Phys. Chem. 1986; 90: 4590. 68. Tarabara VV, Nabiev IR, Feofanov AV. Langmuir 1998; 14: 1092.
J. Raman Spectrosc. 2007; 38: 323–331 DOI: 10.1002/jrs
331
