Relaxation Behaviors of Monolayers of Octadecylamine and Stearic

This study investigated the relaxation behaviors of octadecylamine (ODA), stearic acid (SA), and SA/ODA mixed monolayers at the air/water interface...
5 downloads 0 Views 616KB Size
Download PDF
3180

Langmuir 2004, 20, 3180-3187

Relaxation Behaviors of Monolayers of Octadecylamine and Stearic Acid at the Air/Water Interface Yuh-Lang Lee* and Kou-Liang Liu Department of Chemical Engineering, National Cheng Kung University, Tainan 70101, Taiwan, Republic of China Received July 2, 2003. In Final Form: January 23, 2004 This study investigated the relaxation behaviors of octadecylamine (ODA), stearic acid (SA), and SA/ODA mixed monolayers at the air/water interface. Area relaxations of monolayers at constant surface pressure were studied by a nucleation and growth mechanism and by direct observation using a Brewster angle microscope (BAM). The results showed that ODA and SA monolayers exhibit different characteristics in the area loss and in the BAM morphology. In the initial relaxation stage, SA monolayer illustrates a more stable characteristic than ODA. But at the later stage, the area loss of SA monolayer increases more quickly than that for ODA due to significant nucleation and growth of 3D aggregates. The BAM results demonstrated that 3D aggregates of large scale domains are likely to form on a SA monolayer even when the area loss is insignificant. On the contrary, only dotlike aggregates of low density were found on the ODA monolayer when relaxation is carried out at higher surface pressure. The relaxation behavior of SA monolayer can be described by the Vollhardt model. However, the relaxation of ODA monolayer does not follow the nucleation model described by Vollhard but can reasonably be attributed to the effect of dissolution. For the SA/ODA mixed monolayers, the relaxation behaviors in the initial and final stages follow different mechanisms, which is attributed to the formation of distinct phases as observed from the BAM. This result also implied that SA and ODA are not completely miscible to be a homogeneous phase. Phases of various compositions were formed in the mixed monolayers, and thus, the relaxation mechanism was shifted during the relaxation process as dominated by different relaxation behaviors of various phases.

1. Introduction The Langmuir-Blodgett (LB) deposition technique is known to be capable of preparing highly ordered monomolecular films with densely packed structure and precisely controlled thickness.1,2 Such properties have provided potential applications of LB films in various practical fields.3,4 However, the quality of a LB film is intimately related to the stability and characteristics of the precursor monolayer at the air-water interface.5-7 Thus, many investigations have been dedicated to determining the characteristics and behaviors of monolayers at the air-water interface.8-10 It has been shown that parts of the defects in the LB films are formed already on the monolayer11-13 especially * Corresponding author: e-mail, [email protected]; fax, 886-6-2344496. (1) Langmuir-Blodgett Films; Roberts, G., Ed.; Plenum Press: New York, 1990. (2) Schwartz, D. K. Surf. Sci. Rep. 1997, 27, 241. (3) Kuhn, H.; Mobius, D.; Bucher, H. In Physical Methods of Chemistry, 2nd ed.; Weissenberg, A., Rossiter, V. M., Eds.; Investigation of Surfaces and Interfaces Vol. 9B; Interscience: New York, 1993; Chapter 6, p 375. (4) Proceeding of the 6th International Conference on Organized Molecular Films; Toris-Rivieres, Canada, 1993; Leblanc, R. M., Salesse, C., Eds.; Elsevier: Amsterdam, 1994. (5) Ariga, K.; Shin, J. S.; Kunitake, T. J. Colloid Interface Sci. 1995, 170, 440-448. (6) Brinks, B. P. Adv. Colloid Interface Sci. 1991, 34, 343. (7) Morelis, R. M.; Girard-Egrot, A. P.; Coulet, P. R. Langmuir 1993, 9, 3101-3106. (8) Kajiyama, T.; Oishi, Y. In New Developments in Construction and Functions of Organic Thin Films; Kajiyama, T., Aizawa, M., Eds.; Novel Concepts of Aggregation Structure of Fatty Acid Monolayers on the Water Surface; Elsevier Science B.V.: Amsterdam, 1996; pp 1-38. (9) Vollhardt, D. Adv. Colloid Interface Sci. 1996, 64, 143. (10) Chou, T. H.; Chang, C. H. Colloids Surf., B 2000, 17, 71. (11) Morelis, R. M.; Girard-Egrot, A. P.; Coulet, P. R. Lamgmuir 1993, 9, 3101 (12) Angelova, A.; Penacorada, F.; Stiller, B.; Zetzsche, T.; Ionov, R.; Kamusewitz, H.; Brehmer, L. J. Phys. Chem. 1994, 98, 6790.

when the deposition is performed at a surface pressure above the equilibrium surface pressure (ESP). The majority of defects on the monolayer are formed as a consequence of transformation of two-dimensional (2D) monolayer into three-dimensional (3D) aggregates. When a monolayer is in a supersaturated state at a surface pressure above the ESP, relaxation may occur which leads to nucleation and growth of 3D structures. The homogeneity of the monolayer is thus disturbed and an in-plane discontinuity and heterogeneity of deposited LB film results. The transformation kinetics of a monolayer can be easily accounted for by the decrease of surface area (or surface pressure) during relaxation under constant pressure (or constant area). Nucleation and growth theories have been introduced to describe the area loss of monolayers held at constant surface pressure and to model the mechanism of 2D to 3D transformation during relaxation.14-17 Besides, with the initiation of Brewster angle microscopy (BAM) for monolayer characterization, it has become possible to visualize the inner structure of condense phases directly at the air/water interface.18-21 Such development of BAM also provides the possibility to monitor the dynamics of a transformation process during monolayer relaxation. The characteristics of fatty acids at the air/liquid interface have been widely studied as well as their relaxation behavior at constant surface pressure.17,21-23 (13) Peltonen, J. P.; He, P.; Rosenholm, J. B. J. Am. Chem. Soc. 1992, 114, 7637. (14) Smith, R. D.; Berg, J. C. J. Colloid Interface Sci. 1980, 74, 273. (15) Vollhardt, D.; Retter, U. J. Phys. Chem. 1991, 95, 3723. (16) Wagner, J.; Michel, T.; Nitsch, W. Langmuir 1996, 12, 2807. (17) Vollhardt, D.; Retter, U. Langmuir 1998, 14, 7250. (18) Honig, D.; Mobius, D. Thin Solid Films 1992, 210/211, 64. (19) Honig, D.; Overbeck, G. A.; Mobius, D. Adv. Mater. 1991, 4, 419. (20) Siegel, S.; Honig, D.; Vollhardt, D.; Mobius, D. J. Phys. Chem. 1992, 96, 8157. (21) Angelova, A.; Vollhardt, D.; Ionov, R. J. Phys. Chem. 1996, 100, 10710.

10.1021/la030272q CCC: $27.50 © 2004 American Chemical Society Published on Web 03/11/2004

Relaxation Behavior of Monolayers

However, there is surprisingly little attention paid to the study of monolayers of fatty amines24-27 compared to that of fatty acid. The monolayers of fatty amines have recently been used as an adsorption layer for deposition of nanoparticles,28,29 salts,30,31 clay,32 or enzymes33 from the aqueous phase and, after suitable treatment, used for formation of ultrathin films of nanoclusters and oxides or for biosensor. In addition, mixed monolayers of fatty amines and fatty acids had also been used to improve the stability of monolayer on the acidic subphase and used to prepare two-dimensional arrays of nanoparticles.29 Such applications of fatty amines have raised the importance of the study of the characteristics of amine monolayers as well as its mixed monolayers. In this work, the relaxation behaviors of octadecylamine (ODA), stearic acid (SA), and SA/ODA mixed monolayers were studied in terms of the area loss and the phase images of BAM during the relaxation process. The characteristics of monolayers in the area loss and BAM images are related, and the differences among ODA, SA, and their mixed monolayers were compared. 2. Experimental Section Stearic acid (>99% pure) was purchased from Fluka, and octadecylamine (>99% pure) was supplied by Aldrich, USA. These materials were used without further purification. Chloroform (>99% pure), purchased from J. T. Baker (USA), was used as the spreading solvent to prepare stock solution of concentration 1 mg/mL. The water used was purified by means of a Milli-Q plus water purification system with an electric resistance of 18.2 MΩ. The experiments of monolayer relaxation were performed on a computer-controlled film-balance apparatus with a Teflon trough constructed by Nima Technlogy Ltd, England (model 601 BAM). The trough has a working area of 70 × 7 cm2 and was placed on a vibration isolation table and enclosed in an environmental chamber. The temperature of the subphase was controlled at 25 °C by an external circulator. The surface pressure was measured by the Wilhelmy plate method using filter paper. The resolution for the surface pressure measurement is 0.004 mN/m, according to the instrument specifications. The sample containing monolayer-forming materials was spread on the subphase by microsyringe. After 20 min was allowed for solvent evaporation, the monolayer at the air/liquid interface was compressed at a rate of 1 Å2/(molecule min). After compression of the monolayer to a preset surface pressure, the surface pressure was then kept constant by automatically adjusting the surface area of the trough through the movement of barriers. A relaxation curve was obtained by recording the trough surface area during the relaxation period. A Brewster angle microscope, designed by Nanofilm Technology (NFT), Go¨ttingen, Germany (model BAM2 plus), was mounted directly on the trough to observe the morphological characteristics (22) Vollhardt, D.; Retter, U.; Siegel, S. Thin Solid Films 1991, 199, 189. (23) Vollhardt, D.; Gutberlet, T. Colloids Surf., A 1995, 102, 257. (24) Lee, Y. L. Langmuir 1999, 15, 1796. (25) Bardosova, M.; Tredgold, R. H.; Ali-Adib, Z. Langmuir 1995, 11, 1273. (26) Gunguly, P.; Paranjape, D. V.; Rondelez, F. Langmuir 1997, 13, 5433. (27) Chovelon, J. M.; wan, K.; Jaffrezic-Renault, N. Langmuir 2000, 16, 6223. (28) Muramatsu, K.; Takahashi, M.; Tajima, K.; Kobayashi, K. J. Colloid Interface Sci. 2001, 242, 127. (29) Du, H.; Bai, Y. B.; Hui, Z.; Li, L. S.; Chen, Y. M.; Tang, X. Y.; Li, T. J. Langmuir 1997, 13, 2538. (30) Ganguly, P.; Paranjape, D. V.; Sastry, M. Langmuir 1993, 9, 571. (31) Ganguly, P.; Paranjape, D. V.; Sastry, M. J. Am. Chem. Soc. 1993, 17, 793. (32) Ras, R. H. A.; Johnston, C. T.; Franses, E. I.; Ramaekers, R.; Maes, G.; Foubert, P.; De Schryver, F. C.; Schoonheydt, R. A. Langmuir 2003, 19, 4295. (33) Jin, J.; Li, L. S.; Wang, X.; Li, Y.; Zhang, Y. J.; Chen, X.; Li, Y. Z.; Li, T. J. Langmuir 1999, 15, 6969.

Langmuir, Vol. 20, No. 8, 2004 3181

Figure 1. Relaxation curves for the SA, ODA, and SA/ODA mixed monolayers on the subphase of pure water. The relaxation is controlled at constant surface pressure of 15 mN/m at 25 °C. of the monolayer at the air/water interface. The p-polarized light from an Nd:YAG laser, with a wavelength of 532 nm, was incident at the Brewster angle (53.1°) to the air/water interface. Under such conditions, the reflectivity of the beam was almost zero on the clean water surface. When an insoluble monolayer was present at the air/water interface, the beam reflects and an image was shown according to the organization of the monolayer. The images were visualized on a CCD camera and were recorded by means of a video recorder. The lateral resolution of BAM was about 2 µm. The BAM images were obtained during the relaxation process.

3. Results and Discussion Area Relaxation and the Related Mechanism Predicted by the Theoretical Model. The equilibrium surface pressures (ESPs) for the monolayers of SA and ODA were estimated to be 18 and 9 mN/m, respectively, on pure water. For SA, this value is slightly higher than that reported by Vollhardt and Retter,17 15 mN/m, which was obtained on acidic subphase (pH ) 3). Figure 1 shows the typical relaxation curves for SA, ODA, and SA/ODA monolayers at a surface pressure of 15 mN/m. The SA monolayer exhibits a very stable characteristic with negligible area loss during the relaxation period. On the contrary, the area loss of ODA is about 20% after 120 min of relaxation. The distinction between the two compounds at such condition can be attributed to the difference in ESP. Since the relaxation pressure (15 mN/m) is smaller than the ESP of SA (18 mN/m), the nucleation and growth events on a SA monolayer are considered to be insignificant. On the contrary, because the relaxation pressure is higher than the ESP of ODA (9 mN/m), significant area loss was observed. For the mixed monolayer, the area loss increases with increase of molar ratio of ODA, which is a result of the combination effects of the two components. When the relaxation surface pressure was elevated up to 30 mN/m, a value higher than the ESPs of the two components, the relaxation behavior becomes more complicated as shown in Figure 2. In the early stage of the relaxation process, the SA monolayer also exhibits a more stable characteristic than that for ODA and other mixed monolayers. However, in the later stage, the relaxation rate of SA monolayer increases very quickly and becomes the least stable one with area loss of about 50% within 30 min. For the ODA monolayer, although the area decreasing rate is comparatively higher in the initial stage, the decreasing rate becomes moderate gradually. After 120

3182

Langmuir, Vol. 20, No. 8, 2004

Lee and Liu

Figure 2. Relaxation curves for the SA, ODA, and SA/ODA mixed monolayers on the subphase of pure water. The relaxation is controlled at constant surface pressure of 30 mN/m at 25 °C.

Figure 3. Comparison of the area relaxation results obtained by experiments (the points) to that obtained by best fits of the data to the model described by eq 1 (the solid lines).

Table 1. The Model Parameters of Relaxation of Monolayers at the Air/Water Interfacea

A∞/A0 x Kx (10-3) Rb Sc (10-3)

pure SA

8/2

6/4

0 2.48 0.2 0.999 3.8

0.34 1.46 2.2 0.985 33.4

0.48 1.12 11.6 0.985 23.8

SA/ODA 5/5 4/6 0.49 1.40 2.5 0.992 17.0

0.57 0.83 33.9 0.992 11.4

2/8

pure ODA

0.72 0.72 47.5 0.994 5.4

0.73 0.71 47.5 0.999 0.7

a The relaxation is carried out at constant surface pressure of 30 mN/m and 25 °C. b R, correlation coefficient. c S, standard error.

min of relaxation, the area loss of ODA is about 20% only. For the mixed monolayers, their relaxation behaviors exhibited the combination characteristics of the two compounds in both initial and later stages. In the early relaxation stage, the area loss of the mixed monolayer increased with increasing concentration of ODA, but in the later stage, the area loss increased with increase of SA composition. The relaxation behavior described above indicates that SA and ODA have distinct relaxation characteristics in both initial and later stages. To study the relaxation mechanism, a model proposed by Vollhardt et al. was used in the present work.15,22 In Vollhardt’s model, the area relaxations of insoluble monolayers at constant surface pressure were theoretically described by the transformation of monolayer material into an overgrown 3D phase. The theory is based on the homogeneous nucleation and succeeding growth of the centers, considering simultaneously the nucleation rate, growth rate, and overlapping of the growing centers. A general expression for different mechanisms of nucleation and growth was expressed as

Ao - A ) 1 - exp(-Kxtx) Ao - A ∞

(1)

where Ao is the initial monolayer area, A is the monolayer area at time t, and A∞ is the area of monolayer for t f ∞. The exponent x is a characteristic quantity related to a particular nucleation and growth mechanism specified by the overall transformation constant Kx. The value of A∞ can be estimated by plotting A/Ao values versus 1/t. The intersection obtained by extrapolating the curve to (1/t) f 0 gives a good approximation of the A∞/Ao value.

Figure 4. Constant pressure relaxation data of ODA and SA monolayers expressed as ln(A/Ao) vs t1/2 to examine the dissolution effect of molecules on the area loss. Table 2. The Model Parameters of Relaxation of SA/ODA Mixed Monolayers at Air/Water Interface Obtained in the Initial Stage (0-20 min) and Later Stages (40-120 min)a time interval (min) 0-20

40-120

A∞/A0 x Kx (10-3) Rb Sc (10-3) A∞/A0 x Kx (10-3) Rb Sc (10-3)

8/2

6/4

SA/ODA 5/5

4/6

2/8

0.34 1.59 0.75 0.991 2.1 0.41 0.61 31.06 0.997 3.9

0.48 1.38 3.54 0.992 3.6 0.69 0.72 46.10 0.999 1.8

0.49 0.95 6.85 0.996 1.30 0.59 0.78 43.54 0.994 8.6

0.57 1.27 8.59 0.997 2.8 0.75 0.88 18.32 0.998 1.9

0.72 0.77 37.30 0.999 0.0 0.83 0.70 32.06 0.996 1.8

a The relaxation is carried out at constant surface pressure of 30 mN/m and 25 °C. b R, correlation coefficient. c S, standard error.

By fitting the experimental data of Figure 2 (π ) 30 mN/m) with eq 1, the exponent x of the best fit can be evaluated for each curve. The estimated A∞/Ao values and best-fitted values of x and Kx, as well as the correction coefficients (R) and standard error (S) for various compositions are listed in Table 1. Good agreements between the experimental data and theoretical

Relaxation Behavior of Monolayers

Figure 5. Area relaxation data of the early stage (0-20 min) fitted by the model described by eq 1. A comparison between the experimental data (the points) and the fitting results (the solid lines).

model were observed for pure SA and ODA monolayers, which had a small value for the standard error (S) and a nearly unity value for the correction coefficient (R). However, for the mixed monolayers, the S values are comparatively higher, although the R value is also close to unity. Comparisons between the experimental data and the fitting results are also shown in Figure 3 for monolayers of SA, ODA, and SA/ODA ) 6/4. These results indicate that Vollhardt’s model cannot be applied to the

Langmuir, Vol. 20, No. 8, 2004 3183

long-term relaxation process (120 min) of the SA/ODA mixed monolayers. For the SA monolayer, the exponent x obtained for π ) 30 mN/m is about 2.5. According to Vollhardt’s model,22 the relaxation follows a mechanism of progressive nucleation and hemispherical edge growth. A similar result was reported for SA monolayers spread on acidic solution (pH ) 3) at π ) 34 mN/m and 20 °C.15 For the ODA monolayer, the exponent x obtained is about 0.7. The exact nucleation mechanism with the time exponent x ) 0.7 was not described by Vollhardt et al. or by other models in the literature. The main characteristic in the relaxation of ODA monolayer is the higher area loss in the initial relaxation stage, and thus, the A-t curve shows a concaveupward shape which is contrary to the concave-downward shape of SA. Such relaxation characteristic of ODA was also found on a subphase of alkaline solution24 which exhibited a much more stable property than that obtained on pure water. Another compound which showed a similar characteristic to ODA is dipalmitory lecithin found by Smith and Berg.14 A plausible explanation proposed by Smith and Berg14 for this phenomenon is structure rearrangement within the monolayer to relieve surface inhomogeneities resulting from compression of the monolayer. In addition to the reasons discussed above, the other possibility leading to the area loss of ODA is its high solubility in water which had been reported in the literature.25-27 According to a model proposed by Smith and Berg,14 when the area loss of monolayer was caused from the dissolution of molecules into the subphase

Figure 6. BAM images of SA (a) and ODA (b) monolayer relaxation at π ) 15 mN/m. (a-1, b-1) show images after compression to 15 mN/m. The length bar shown in the figure corresponds to a length of 50 µm.

3184

Langmuir, Vol. 20, No. 8, 2004

through diffusion, a linear relationship would be obtained between ln(A/Ao) and t1/2. This relationship was examined and is shown in Figure 4. The linear relationship of ODA data did imply the importance of the dissolution effect of ODA on the area loss. Since the theoretical model cannot give a satisfactory prediction to the entire relaxation process (120 min) of the mixed monolayer, as shown in Figure 3, it is quite possible that a more complicated relaxation process occurred in the mixed monolayer and that the dominating mechanism in the early and later relaxation stages is different. To verify this inference, relaxation data in the early stage (0-20 min) were used to get the best fitting of the theoretical model, and the results are listed in Table 2. Good consistence between the theoretical model and experimental data was inspected from Figure 5, which can also be verified by the small value of standard error (S) shown in Table 2. The relaxation behavior in the later stage (40-120 min) also showed good consistence between the experimental data and the theoretical model as examined by the small S value listed in Table 2. The exponent x in the later stage is smaller compared with that of the early relaxation stage, especially for the monolayers with higher composition of SA. This result sustains the previous inference that the relaxation of the mixed monolayer follows different mechanisms in the initial and later stages. In the early stage, the relaxation behavior of mixed monolayer exhibits a combination result which depends on the composition of the mixed monolayer. However, in the later stage, the relaxation mechanisms do not vary significantly with variation of SA composition and their time exponents resemble that of the pure ODA monolayer. The analysis discussed above exhibits that relaxations of SA and ODA monolayers can be correctly fitted by eq 1 although a specific nucleation model cannot be assigned to the time exponent of ODA. For the long time relaxation (120 min), the mismatch between the relaxation data of mixed monolayer to eq 1 may imply that SA and ODA are not completely miscible. Phases of various compositions may coexist in the mixed monolayer which render the various relaxation mechanisms in the initial and later stages, respectively. Such inference is sustained by the observations of BAM described in the later section. Relaxation Behavior Observed by Brewster Angle Microscopy. The BAM images were obtained simultaneously in the area relaxation experiments of the monolayers. Figure 6 shows the BAM images of SA and ODA monolayers before and after relaxation at π ) 15 mN/m. When the monolayers were compressed to π ) 15 mN/m, homogeneous reflecting images were observed on SA (Figure 6(a-1)) and ODA (Figure 6(b-1)) monolayers. However, on the SA monolayer (Figure 6a), patchlike domains of homogeneous structure can be visible, although with low contrast. During the compression process, separated domains are also visible on the SA monolayer (not shown here). Such domains move close and finally contact each other upon compression. After further compression, the contrast of the separated domains becomes lower or, even more, invisible. So, it is quite possible that the separated domains on the SA monolayer correspond to phases of various packing structures such as liquid expanded and liquid condensed phases. During the relaxation process, the separated domains did not disappear and little bright dots appeared on the phase boundary as shown in Figure 6(a-2) after 120 min of relaxation. The bright dots correspond to the 3D aggregates initiated by the mismatch structures of different SA domains. Despite

Lee and Liu

Figure 7. BAM images of SA/ODA (1/1) mixed monolayer relaxation at π ) 15 mN/m: (a) after compression to 15 mN/m; (b) after a relaxation time of 120 min. The length bar shown in the figure corresponds to a length of 50 µm.

the formation of 3D aggregates, the area loss in the relaxation process is negligible during 120 min (A/A0 ) 0.996). On the other hand, the BAM images of ODA monolayer show no difference during the relaxation period although the area loss is about 20% after 120 min of relaxation. This phenomenon implies that if 3D aggregates formed in the relaxation process, the aggregates should be smaller than a micrometer, which cannot be visible by BAM. However, the dissolution effect should take an important part in the area loss as discussed in the previous section. That is, if the area loss during the relaxation is caused by dissolving ODA molecules into the subphase but not by the formation of 3D aggregates, it is reasonable that the BAM images will keep a stable and homogeneous phase. For the mixed monolayer, patchlike domains were also visible in the compression process and, furthermore, bright dots were found on the boundary of these domains. It is very possible that the formation of such bright dots is also caused from the mismatch structure of separated domains as those on the SA monolayer, but the existing of ODA molecules did play an enhancing effect to the formation of new phase or 3D aggregates. On further compression of the monolayer, the bright dots grow and coalesce to curves which seem to resemble the boundary of domains before coalescence as shown in Figure 7a. The BAM images of the mixed monolayer did not vary significantly during the relaxation process at π ) 15 mN/m as shown in Figure 7b.

Relaxation Behavior of Monolayers

Langmuir, Vol. 20, No. 8, 2004 3185

Figure 8. BAM images of SA (a) and ODA (b) monolayers relaxation at π ) 30 mN/m. (a-1, b-1) show the images after compression to 30 mN/m. The length bar shown in the figure corresponds to a length of 50 µm.

BAM Images at Higher Relaxation Pressure. When the relaxation pressure was elevated to 30 mN/m, more aggregated structures can be observed as shown in Figure 8. When SA monolayer was compressed to π ) 30 mN/m, obvious bright dots had already formed on the monolayer (Figure 8(a-1)). During the relaxation process, the aggregates grow quickly and form a platelet structure with a size of hundreds of micrometers after 7 min of relaxation (Figure 8(a-2)). However, the area loss at this relaxation time is only 3%. In addition to the large platelike aggregates, small dots of various sizes were also visible in this figure. The wide dispersion of the size of the aggregates on SA monolayer sustains the inference of the

relaxation model that the relaxation of SA monolayer follows a mechanism of progressive nucleation as described in the previous section. The area of aggregates on BAM image increases gradually with increase of relaxation time and approaches a ratio over 50% after 25 min of relaxation (Figure 8(a-3)). For ODA monolayer, no visible aggregate was observed before relaxation (Figure 8(b-1)). After 7 min of relaxation, the area loss is about 5% and little bright dots of micrometer size appeared (Figure 8(b-2)). With further relaxation, the density of bright dots increases slightly but the size of the aggregates did not increase significantly even after 120 min of relaxation (Figure 8(b-3)).

3186

Langmuir, Vol. 20, No. 8, 2004

Lee and Liu

Figure 9. BAM images of SA/ODA mixed monolayers relaxation at π ) 30 mN/m. The compositions XSA are 0.6 (a) and 0.4 (b). (a-1, b-1) show the images after compression to 30 mN/m. The length bar shown in the figure corresponds to a length of 50 µm.

The BAM images of SA/ODA mixed monolayers relaxed at π ) 30 mN/m are shown in Figure 9 for XSA ) 0.6 and 0.4. After compression to π ) 30 mN/m, two types of domains formed on the mixed monolayers. The first one exhibits a dark and homogeneous image without significant aggregates, but the other contains densely distributed bright dots. For the second domain, the mixed monolayer with higher composition of SA, XSA ) 0.6, illustrates a higher density of bright dots (Figure 9(a-1)) than that of the monolayer with smaller SA composition, XSA ) 0.4, shown in Figure 9(b-1). During the relaxation process, the bright domain patches can also be observed in the mixed monolayers, as show in parts a-2 and b-2 of Figure

9, which are supposed to be caused from the growth and coalescence of the dispersed dotlike aggregates. Since bright dots and patch domains were easy to form on a SA monolayer, it is very likely that the bright dots were formed as a result of SA molecules which dispersed in a ODA monolayer. This inference also implies that SA and ODA are not completely miscible on the gas-liquid interface. It is more likely that one compound dispersed in the other rather than forming a homogeneous phase. On the other hand, the darkly homogeneous image indicates a more homogeneous phase of this region and its relaxation behavior is similar to that of the ODA monolayer.

Relaxation Behavior of Monolayers

The distinct relaxation images of the two regions is consistent with the previous analysis on the area loss which concluded that relaxation of the mixed monolayer follows different mechanisms in the initial and later stages of constant pressure relaxation. In the initial stage, the relaxation process was mainly attributed to the region with densely distributed aggregates. The composition dependence of time exponent x in this stage, shown in Table 2, is verified by the distinction of dot density between parts a and b of Figure 9. In the later stage, the values of time exponent x for the mixed monolayers were similar and resemble that of ODA as shown in Table 2. This result suggests that the relaxation characteristic and composition of the homogeneous region in the mixed monolayers were also similar to ODA and its relaxation behavior is the dominating step in the later relaxation stage. For the monolayer with composition XAS ) 0.6, the reflecting intensity from the monolayer decreases gradually with further increase of the relaxation time, and thus, the brightness of the image becomes smaller and the contrast between various domains becomes lower. For a relaxation time of 110 min, the area ratio decreases to 0.56 and the respected image is shown in Figure 9(a-3). However, the situation seems different on the monolayer with higher composition of ODA (XSA ) 0.4). As shown in Figure 9(b-3), which also has a relaxation time of 110 min but with smaller area loss (A/A0 ) 0.67), the dotlike aggregates grow larger without decreasing the brightness and contrast of the images. From the BAM results shown above, we may conclude that 3D aggregates of micrometer scale easily form on a SA monolayer, especially at high surface pressure. On the contrary, large scale aggregates are not visible on the ODA monolayer. Such results exhibit a consistent relationship to the surface wettability of LB films of SA and ODA shown in previous papers.24,34 The advancing contact angle of water on one layer of ODA LB film deposited on hydrophilic glass is about 114°, which is equivalent to a methyl-terminated surface. However, the contact angle on a SA LB film, prepared at similar conditions as for ODA, is much smaller and less stable, which reflects the (34) Lee, Y. L.; Chen, C. Y. Appl. Surf. Sci. 2003, 207, 51.

Langmuir, Vol. 20, No. 8, 2004 3187

irregular packing and orientation of SA molecules. Because the uniformity and quality of a LB film are strongly affected by the stability of monolayer, the BAM results obtained in this work give a reasonable explanation to the difference of wettability between SA and ODA LB films. 4. Conclusions Area relaxation behaviors of SA, ODA, and mixed SA/ ODA monolayers were studied. The results showed that SA and ODA monolayers exhibit different characteristics both in the area loss and in the BAM morphology during the relaxation process. The SA monolayer is very stable in the initial relaxation stage, but the area loss increases quickly in the later stage due to significant nucleation and growth mechanism. According to Vollhardt’s model, the relaxation of SA follows a mechanism of progressive nucleation and hemispherical edge growth. For the ODA monolayer, the initial area loss is comparatively higher than that of the SA monolayer, but the area loss increases moderately with relaxation time and exhibits a more stable characteristic in the later stage. The relaxation behavior of ODA cannot be described by the nucleation and growth mechanism of Vollhard’s model but can be reasonably attributed to the effect of dissolution. The results of the BAM observation demonstrate that 3D aggregates of large scale domains are likely to form on a SA monolayer even when the area loss is small. On ODA monolayer, only dotlike aggregates of low density were formed when relaxation is carried out at higher surface pressure. For the SA/ODA mixed monolayer, the relaxation behaviors in the initial and final stages follow different mechanisms, which can be attributed to the formation of distinct phases due to the incomplete miscibility of SA and ODA molecules on the gas/liquid interface. Such inference is sustained by the observation of BAM. Acknowledgment. The support of this research by the National Science Council of the Republic of China through Grant No. NSC 92-2214-E-006-024 is gratefully acknowledged. LA030272Q