Influence of Single versus Double Hydrogen-Bonding Motif on the

Both showed IR absorptions corresponding to the hydrogen-bonded N−H and C O groups at ≅ 3324 and 1682 cm-1, respectively. This indicates that the ...
0 downloads 0 Views 1MB Size
J. Phys. Chem. B 2006, 110, 15251-15260

15251

Influence of Single versus Double Hydrogen-Bonding Motif on the Crystallization and Morphology of Self-Assembling Carbamates with Alkyl Side Chains: Model System for Polyurethanes Shalini Khanna, Mohammad Moniruzzaman,† and Pudupadi R. Sundararajan* Ottawa Carleton Chemistry Institute, Carleton UniVersity, 1125 Colonel By DriVe, Ottawa, Ontario K1S 5B6, Canada ReceiVed: March 30, 2006; In Final Form: June 7, 2006

The difference in the morphology and crystallization aspects of hydrogen-bond-mediated self-assembling systems with single and double hydrogen-bonding motifs is studied here with carbamates as an example. These carbamates have alkyl side chains of various lengths, from C4 to C18. The biscarbamates with double hydrogen-bonding sites and symmetric substitution of alkyl segments show a significantly different morphological behavior as compared to the N-octadecyl carbamate alkyl esters (ref 5, referred to as simple carbamates henceforth) with a single hydrogen-bond motif and asymmetric substitution of alkyl side chains. In contrast to the simple carbamates in which no significant difference was found in the spherulite size from C4 to C12, with the biscarbamates we find that the spherulitic size, rate of growth of spherulites, and rate of crystallization show a maximum with an alkyl chain length of C8. This is rationalized in terms of the relative contributions of the hydrogen-bond and van der Waals interaction energies. Oriented X-ray diffraction patterns from the fibrils of the spherulites lead to a model for the growth patterns of the hydrogen-bond planes and the molecular orientation in the spherulites.

Introduction Hydrogen-bond-mediated self-assembly has been discussed extensively to create supramolecular architectures and transient polymer systems.1 Such self-assembled complexes leading to dimerization or polymerization have been studied in solution and in the solid state.2 Hydrogen-bonded supramolecular polymer networks in the solid state have also been reported.3 It has been noted that the strength and selectivity of such hydrogen bonds can be increased by introducing arrays of donor and acceptor groups, leading to double or triple hydrogen-bonding sites. Self-complimentarity is achieved in such cases. It has been shown that extended supramolecular assembly requires such multiple complementary hydrogen bonds.1 Single hydrogenbonding motifs have been used to design transient liquid crystalline polymers.4 While hydrogen bonding would lead to supramolecular assembly, one should be able to control the morphology of such assembly by, e.g., choice of side groups. Formation of single crystals, spherulites, or gels can occur depending on the conditions of solidification. If the process, for example, leads to large spherulites, the resultant solid would be opaque and brittle. It thus becomes important to tailor the molecule to achieve the morphology to match the desired functional property. We have been studying the influence of side chains on the morphology of a class of self-assembling system of carbamates5,6 and biscarbamates. Carbamates and biscarbamates, as a class of compounds, have a long history in terms of scientific studies and technological applications. These are considered model compounds for * To whom correspondence should be addressed. E-mail: Sundar@ Carleton.ca. † Present Address: Department of Materials Science and Engineering, University of Pennsylvania, Philadelphia, PA 19104-6272.

polyurethanes. Acetylene biscarbamates were investigated by Gaylord7 in the early 1950s. Carbamates and biscarbamates have also been used as organic intermediates.8 The decomposition mechanism of polyurethanes has been discussed in terms of that of biscarbamates.9 They have been used to adjust the viscosity of oil and grease and improve hardness of polyurethane-based adhesive and sealants10-13 and as antiinflammatory agents,14 pesticides,15 and sequence-specific DNA alkylation agents.16 These have also been investigated as photosynthetic inhibitors.17 Because of the possibility of hydrogen bonding, the association process in the case of carbamates is similar to that of biological self-assembly. Studies on hydrogen bonding and IR spectra of various carbamates have been reported by Furer.18 Molecular organization of mono- and dicarbamates on surfaces was discussed by Matzger and Kim19 to discern the mechanism of adhesion of polyurethanes. Application of carbamates with alkyl side chains have been investigated for use as ink vehicles for ink-jet printing technologies.20-22 These carbamates have a melt viscosity of about 8-12 centipoises and melting temperatures of 60-90 °C and crystallize rapidly upon quenching from the melt. These attributes are suitable for printing ink technologies. In a previous publication5 we studied the morphology and thermal behavior of N-octadecyl carbamates (we will refer to these as simple carbamates henceforth) with alkyl side chains of various lengths from C4 to C18. These are molecules with a single hydrogen-bonding motif and asymmetric disposition of alkyl chains on either side of the hydrogen-bonding motif (except in the case of C18). It was found that the size of the spherulites changed moderately (∼50-200 µm) with the length of the alkyl side chain from C4 to C12, but it was an order of magnitude larger (∼1200 µm) with C18 (symmetric) side chain. Significant differences were also found between the morphologies upon quenching from the melt and slow cooling of the

10.1021/jp0619843 CCC: $33.50 © 2006 American Chemical Society Published on Web 07/14/2006

15252 J. Phys. Chem. B, Vol. 110, No. 31, 2006 sample. The melting temperature increased from 61 to 73 °C as the side chain length increased from C4 to C12, and it was 87 °C for C18. It was shown that the morphology in each case can be modified by blends of homologous carbamates. It was also found that the extent of hydrogen bonding was not affected by blending any two carbamates to reduce the crystallinity and spherulite size, while the packing of the alkyl chain was. We also discussed the effect of nucleating agents and clarifiers on controlling the spherulite size and crystallinity of these simple carbamates23 and the role of the alkyl chain length and solvent on their capability to form organogels.6 In this work, we discuss the morphology of carbamates with double hydrogen-bonding motifs, with symmetric alkyl substitution on either side of the hydrogen-bonding groups, and compare it with the simple carbamates with a single hydrogen-bonding moiety. The difference in the sequence and the symmetry of the alkyl side chains on either side of the hydrogen-bonding motif as illustrated below for a C8 side chain would no doubt affect the morphology and crystallization characteristics.

Khanna et al.

Figure 1. Melting (Tm) and crystallization temperatures (Tc) of slowcooled and quenched samples for biscarbamates with different alkyl chain lengths. The Tm’s of the simple carbamates are also shown.

flow curve with respect to time. The fractional degree of crystallization (R) can be represented by the ratio of the heat of crystallization at time t as

∆Ht R) ) ∆H∞ Both are hydrogen-bonding systems. It is to be noted that in the case of the simple carbamates the octadecyl isocyanate was reacted with different alcohols ranging from 1-butanol to 1-octadecanol, which led to asymmetric alkyl chain lengths on either side of the hydrogen-bonding moiety, except in the last case (C18). In the present case, 1,6-diisocyanatohexane was reacted with various alcohols, leading to symmetric lengths of the alkyl chains. To our knowledge, this is the first time the crystallization kinetics of these materials has been reported. Experimental Section Synthesis of the biscarbamates was performed as described by Goodbrand et al.24 This is briefly described in the Supporting Information. A DuPont 2010 differential scanning calorimeter was used for thermal analysis with a heating rate of 10 deg/min. The instrument was calibrated for temperature and energy with indium and tin reference samples. The samples were run in a nitrogen atmosphere. DSC thermograms were recorded using the Thermal Advantage software, and data were analyzed by TA Universal Analysis. Samples for the morphological studies were prepared in different ways for DSC and microscopy measurements: (1) The ‘slow-cooled’ sample was prepared by melting the material at a temperature of 20 °C higher than its melting point and held for about 15 min to remove any morphological history. It was then cooled slowly to room temperature at a rate of 10 deg/ min. (2) The ‘quenched’ samples were prepared by following the same procedure except that they were quenched immediately from the melt. (3) Isothermal crystallization (at Tc) was also performed for studying the kinetics of crystallization and spherulite growth rate. Isothermal crystallization kinetics was studied using a TA Instruments Q100 MDSC. After melting the samples in the DSC, they were cooled to their crystallization temperature and the heat of crystallization was measured at intervals of 0.60 s. The crystallinity was determined by integrating the isothermal heat

dt ∫t 0 dQ dt

∫∞

0

dQ dt dt

(1)

Here ∆Ht is the partial area between the DSC curve and the time axis at time t and ∆H∞ is the total area under the peak that corresponds to the overall heat of crystallization. The t1/2 was calculated by plotting the fractional degree of crystallization (R) with time (t). A Zeiss Axioplan polarized optical microscope (OM) was used for recording optical micrographs. A Linkam LTS 350 hot stage with a Linkam TMS 94 controller was used to study the kinetics of spherulite growth. The samples were heated at a rate of 5 °C/min to a temperature of 20 °C higher than their melting point and held there for about 10 min. They were then cooled at a rate of 2 °C/min up to 5 °C above the crystallization temperature (Tc) and then at 0.1 °C/min up to the Tc. Northern Eclipse (version 6.0) image processing software was used to capture the images and calculate the spherulite size. A Philips automated powder diffractometer (PW1710) was used for X-ray diffraction studies with Ni-filtered Cu KR radiation (λ ) 1.5418 Å.). Diffraction data were recorded at room temperature, in the range of 2° e 2θ e 50°, using the software MDI Datascan 3.2 (Materials Data Inc., Livermore CA). The results were analyzed using the MDI Jade 5.0 XRD pattern processing software. Samples were prepared in the same way as for the DSC except that in this case glass slides were used to hold the samples. Percent crystallinity (Xc) of each sample was calculated using the relationship

Xc ) (crystalline area under the peak/total crystalline + amorphous area) × 100 (2) Since significant X-ray reflections occur at 2θ < 5°, patterns were also recorded on film using a Statton-type flat film camera (William Warhus Co., Wilmington, DE). Crystallite sizes corresponding to d spacings of 4.6 and 3.8 Å were calculated using the Scherrer equation.25 Fourier transform infrared (FTIR) spectroscopic measurements were performed at ambient conditions using a Michelson

Self-Assembling Carbamates with Alkyl Side Chains

J. Phys. Chem. B, Vol. 110, No. 31, 2006 15253

TABLE 1: Biscarbamates Synthesized from Reaction of an Appropriate Alcohol with 1,6-Diisocyanato Hexane parent molecule

sample ID

molecular formula

molecular length,a Å

Tm, °C (from DSC)

1-butanol 1-hexanol 1-octanol 1-dodecanol 1-hexadecanol 1-octadecanol

C4 C6 C8 C12 C16 C18

H9C4COONH (CH2)6 NHCOOC4H9 H13C6COONH (CH2)6 NHCOOC6H13 H17C8COONH (CH2)6 NHCOOC8H17 H25C12COONH (CH2)6 NHCOOC12H25 H33C16COONH (CH2)6 NHCOOC16H33 H37C18COONH (CH2)6 NHCOOC18H37

23.0 28.1 33.3 43.8 53.5 59.1

91.3 97.1 106.6 114.9 117.8 120.4

a

Length of the molecules calculated using Hyper Chem Pro 6 molecular visualization and simulation software (Hypercube Inc.).

Figure 2. Variation of the heat of fusion with alkyl side chain length of biscarbamates.

M120 BOMEM FTIR spectrometer. The BOMEM GRAMS/ 386 software was used for data collection. Samples were prepared as KBr pellets. Molecular modeling was performed with HyperChem Ver. 6 molecular visualization and simulation software (Hypercube Inc.) with default parameters. Results and Discussion Thermal Behavior. The biscarbamates studied here showed sharp melting points and only one phase transition from solid to the melt. These results along with the absence of any polymorphic transitions prior to melting indicate their purity. The melting temperatures (Tm) along with the crystallization temperatures (Tc) for both the slow-cooled and quenched samples are shown in Figure 1 as a function of the number of carbon atoms in the alkyl side chain. Tm increases sharply from C4 to C8 and then moderately with a further increase in the length of the alkyl side chain. No difference was noted in the melting points of slow-cooled and quenched samples. Tm’s corresponding to the simple carbamates studied before are also shown in this figure for comparison. For each of the side chain lengths the Tm for the biscarbamate is higher than that of the simple carbamate. This can be attributed to both the symmetry of the side chains on either side of the hydrogen-bonding moiety as well as the possibility of two hydrogen bonds for each molecule of the biscarbamate. It is seen from Table 1 and Figure 1 that the increment in Tm decreases with an increase in the alkyl chain length. There is an increase of 9.5 °C from C6 to C8 but only 2.7 °C between C16 and C18. Studies by McKierman et al.26 on aliphatic polyurethanes showed that with an increase in the length of the aliphatic sequence, Tm of the polymer decreased and approached that of polyethylene, due to the reduced contribution of the hydrogen-bond energy, relative to the van der Waals interaction. A similar trend is observed in the present case of the biscarbamates as well. The heat of fusion increases with the alkyl side chain length (Figure 2) with a sharp

Figure 3. Variation of the (a) molecular length (l) and d spacing with the highest relative intensity with the alkyl chain length of biscarbamates, (b) crystallite size (L) corresponding to the 3.8 and 4.6 Å reflections, and (c) relative intensities of the 4.6 (quenched) and 3.8 Å reflections (slow-cooled) samples.

increase from C4 to C8. Similar to the trend in Tm, the rate of increase of heat fusion decreases with the alkyl chain length. As expected, the heats of fusion of slow-cooled samples were found to be higher than those of the quenched samples.

15254 J. Phys. Chem. B, Vol. 110, No. 31, 2006

Khanna et al.

TABLE 2: Interplanar Spacings, d (Å), and Relative Intensities, I (%), Obtained from X-ray Diffraction Patterns of Quenched Samples C4

C6

d

I

33.67 27.24 23.43

w

20.89 19.19

m 100 (s)

10.57 9.98 9.58

6.31

4.62 4.26

3.78

3.31

C8

d

I

2.3 70.5

22.8

100 (s)

17.01

w

13.84

w

11.47

16.7

m w 46.70

2.40

16.03 40.30

59.00

8.49 7.6

w 1.4

6.43 6.10 5.11 4.95 4.61 4.31 4.18

w w w w 28.4 9.9 58.4

3.99 3.80

27.4 20.2

3.72

10.7

3.42

7.5

3.29

3.2

C12

d

I

33.64

4.1

26.91

100 (s)

13.42 13.24

C16

d

I

33.69 27.50

100 (s) 24.8

16.47

w

3.3 m/w

9.85

w

9.44

w

8.87 7.82 7.39 6.7

w w w 7.7

8.41

6.9

4.90 4.65 4.40 4.22 4.12 3.97 3.82

3.67 3.48

6.72

w

w 31.1 22.4 19.7 77.1

5.18 4.95 4.69 4.49 4.29 4.19

w w 45.1 38.6 22.6 16.1

w 58.7

3.81

69.3

14.6 19.8

3.61 3.55

31.7 23.1

2.44

3.7

C18

d

I

39.77

100 (s)

d

I

43.69

100 (s)

27.67

45.1

14.91 13.47

w w/m

14.86

w

11.25 10.13

w 3.6

11.15

4.4

8.18 7.53

w w

7.46

6.0

6.76

3.6

4.99 4.64 4.51 4.35 4.12 4.03

w 62.3 77.6 21.0 23.3 w

5.14 4.98 4.66 4.53 4.39 4.19 4.04

w w 77.1 98.2 11.6 21.6 16.6

3.89

37.9

3.89 3.81

49.6 61.2

3.79 3.58

56.8 64.1 3.60

87.7

5.40

X-ray Diffraction and Structural Features. All the biscarbamate samples exhibit the most intense reflection at 2θ < 5°. Similar to the case of simple carbamates, a number of reflections are seen in the range 18° < 2θ < 26°. The d spacings corresponding to the quenched and slow-cooled samples are listed in Tables 2 and 3, respectively, along with the relative intensities. The large d spacings that could not be recorded with the diffractometer were collected on films, and the intensities of these are denoted as strong (s), medium (m), and weak (w). A crystallographic study has shown that the compound C12 crystallizes in the triclinic crystal system.27 Due to the symmetry of the side chains on either side of the hydrogen-bonding motif, the C12 biscarbamate shows a center of symmetry in the crystal structure. It is not our objective here to determine the crystal structure. However, some comments can be made on the powder diffraction data. It is seen from Table 2 that the largest d spacing recorded for each biscarbamate generally increases from 23.43 Å for C4 to 43.69 Å for C18. Although the largest d spacing is 33.6 Å for C6, C8, and C12, this reflection is most intense only for the latter. It is also noted that the d spacing of the most intense reflection increases from 19.19 to 43.69 Å as the alkyl chain length increases from C4 to C18. Figure 3a shows the linear increase in the calculated length of the molecules and the d spacing of the most intense reflection as a function of chain length.

2.42 2.39

6.6 3.7

Table 2 also shows that a reflection with a d spacing of 4.6 Å (2θ ) 19.29°) is common to all the samples and that its intensity increases almost linearly from C4 to C18 as shown in Figure 3c. To account for this reflection, simple molecular modeling was performed (using the Hyperchem software) by placing four molecules of C12 biscarbamate in proximity and minimizing the energy. The resulting molecular packing is shown in Figure 4. The distance between the alkyl chains in this model is 4.6 Å. This d spacing thus corresponds to the plane of hydrogen bonding, the intensity of which increases with the alkyl chain length (Figure 3c). This reflection was either weak or absent in the case of simple carbamates. The reflection with a spacing of 3.8 Å (2θ ) 23.4°) is also common for the biscarbamates, with a fairly strong intensity. In the previous study on the simple carbamates this reflection was identified with the hydrogen-bonding plane.5 However, it appears that with the biscarbamates, which have double hydrogen bonding for each molecule, the distance between the alkyl chains increases to 4.6 Å. The increase in its intensity with the length of the alkyl chain indicates that side chain packing is enhanced. For this set of molecules, we thus identify the 4.6 Å reflection with the hydrogen-bonding plane and the 3.8 Å reflection with the distance between these planes. The intersheet distance was 3.8 Å in the crystal structure of C12 also.27 It is known that the distance between the chains in the polyethylene crystal structure

Self-Assembling Carbamates with Alkyl Side Chains

J. Phys. Chem. B, Vol. 110, No. 31, 2006 15255

TABLE 3: Interplanar Spacings, d (Å), and Relative Intensities, I (%), from X-ray Diffraction Patterns of Slow-Cooled Samples C4

C6

d

21.06 18.94

I

I

28.34

m

23.23

100 (s)

16.65

w

C12

d

I

30.34

m

26.11

100 (s)

13.18

7.30

I

33.69

100 (s)

C18

d

I

d

I

40.48

100 (s)

43.69

100 (s)

14.86

1.80

13.43

w 11.09

3.400

10.15

10.10 7.39

4.60

5.55

w

4.66 4.53

43 58

4.39

9.30

w 15.2

m 59.70 7.67 7.1 6.33

w w w

5.76

w

6.63

13.50

5.16

w

5.6

w

5.08

2.0

4.62

19.50

4.77 4.64

w 5.30

4.66

w

4.39 4.25

w 52.70

4.41 4.34 4.21

w m 12.0

4.04 3.92 3.82

m w 57.60

3.80

75.30

3.42

6.40

3.45

22.20

4.18

3.77

C16

d

m 100 (s)

12.72 11.59 10.47 9.50

C8

d

7.90 7.77

m w

6.30

w

5.73

w

5.20 5.11 4.9 4.71 4.64

w w w 2.6 w

4.4

m

4.2

7.39

w

6.78 5.80

9.70 w

5.24

w

4.74 4.67

1.3 w

m

4.3 4.21

w w

4.04

w

4.05

7.0

4.19 4.04

10.0 14.20

3.80

53.80

3.80

48.0

3.80

41.90

3.52

34.50

3.57

32.80

3.59

50.30

3.20

45

is about 4.5 Å. A reflection with this spacing was also observed in the case of aliphatic polyurethanes and attributed to the distance between the hydrogen-bonded chains.26 The intersheet distance between such hydrogen-bonded chains was associated with a reflection at 3.7 Å. The crystallite size corresponding to the 3.8 and 4.6 Å reflections (2θ ) 23.4° and 19.3° respectively) for the quenched samples is shown in Figure 3b. The crystallite size for d ) 4.6 Å increases sharply from C4 to C6 with no significant increase thereafter. However, for d ) 3.8 Å, the crystallite size increases sharply from C4 to C8 and decreases significantly with a further increase in the side chain length. This behavior is similar to the spherulitic growth (see below). Table 3 lists some of the reflections for the slow-cooled samples. The 4.6 and 3.8 Å reflections are common to all the biscarbamates. However, the relative intensity recorded for a d spacing of 3.8 Å increases from C4 to C8 and then decreases significantly (Table 3 and Figure 3c) while the intensities for a d spacing of 4.6 Å do not show any trend. This was traced to the molecular orientation (see below) of the lamellae in the spherulites in the samples prepared by slow cooling. Similar to the heat of fusion (Figure 2), X-ray crystallinity increases with an increase in alkyl side chain length. This is not shown here due to uncertainties in the intensities arising from the molecular orientation of the samples Spherulitic Morphology. Cross-polarized optical micrographs for the quenched and slow-cooled samples are shown in Figure 5. As expected, the spherulites from slow-cooled

samples are larger than those obtained by quenching. This is similar to the case of simple carbamates. There is a striking feature in this figure: with both quenching and slow cooling, the spherulite size increases up to C8, and then there is a significant decrease. This departs from the behavior of simple carbamates with which there was no significant difference in spherulite size up to an alkyl chain length of C12, beyond which the size increased significantly. The quenched and slow-cooled samples of C4 did not show any proper spherulites. However, very small birefringent units were observed in the quenched sample, whereas dendrites were seen with the slow-cooled samples. The quenched samples of C6 show fibrillar spherulites with high birefringence. The slow-cooled samples of the same material showed large spherulites with weak birefringence. The optical micrographs of both quenched and slow-cooled samples of compounds C8-C18 showed the presence of spherulites with negative birefringence. This indicates that similar to the case of polyethylene, the alkyl chains are disposed along the tangential direction of the spherulite. Figure 6 shows the variation of spherulite size for quenched and slow-cooled samples as a function of the number of carbon atoms in the alkyl side chain. The spherulite sizes of the slowcooled samples are several times larger than those of the quenched counterparts. As noted above, in both cases spherulite size increases from C6 to C8 and then decreases. Formation of large spherulites is a result of fewer nuclei. The decrease in the size of spherulite with the increase in the length of the alkyl side chain beyond C8 can be attributed to the increase in van

15256 J. Phys. Chem. B, Vol. 110, No. 31, 2006

Khanna et al.

Figure 4. Packing model of biscarbamate with C12 side chain: (a) hydrogen-bonding plane, (b) stacking of hydrogen-bonding planes, (c) schematic of meridional and equatorial reflections in the X-ray diffraction, and (d) sample area used for c (see also Figure 11).

der Waals attraction. This increased association between the alkyl chains could contribute to increased nucleation density. The trend seen above for the variation of the spherulite size with the alkyl chain length can be rationalized on the basis of infrared spectra. No noticeable differences were observed in the IR spectra of slow-cooled and quenched samples. Both showed IR absorptions corresponding to the hydrogen-bonded N-H and CdO groups at = 3324 and 1682 cm-1, respectively. This indicates that the extent of hydrogen bonding is independent of whether the sample was slow-cooled or quenched. However, it is seen from Table 4 that the peak intensity of H-bonded N-H and CdO bands increase slightly from C4 to C8 and then decrease with further increase in the alkyl side chain length. This table shows that the peak width at half-maximum also exhibits a similar trend. Such reduction in the half width was related to the increase in crystalline order in the case of a simple polyurethane.28 The increase in the van der Waals interactions (i.e., attractive contribution to the energy) with an increase in the alkyl side chain length is reflected in the frequencies of the CH2 symmetric (νs) and asymmetric (νas) absorption bands. Figure 7 shows the variation of νs and νas CH2 stretching vibration modes for the different alkyl chain lengths. νs and νas decrease with an increase in the length of the alkyl side chain. For C4, νas and νs are observed at 2935 and 2863 cm-1, respectively. They shift to 2917 and 2849 cm-1 for C18. The shift to lower wavenumbers

Figure 5. Optical micrographs of (left) quenched and (right) slowcooled samples of biscarbamates with different alkyl chain lengths.

from C4 to C18 is indicative of strong organization of the alkyl groups via van der Waals interactions.29 It was noted above that the half width of the peaks of H-bonded N-H and CdO peaks was also found to increase from C4 to C6 and then decrease. The reduction in the half width is indicative of the increase in order.28 When the alkyl side chain is short (C4-C8) the hydrogenbonding interaction dominates over the van der Waals interac-

Self-Assembling Carbamates with Alkyl Side Chains

J. Phys. Chem. B, Vol. 110, No. 31, 2006 15257

Figure 6. Radii of the spherulites of (a) quenched and (b) slow-cooled samples with different alkyl side chain lengths.

Figure 7. Wavenumber (cm-1) of asymmetric and symmetric CH2 stretching vibrations for various alkyl side chain lengths.

tions. As the side chain length increases (C12-C18), the van der Waals attraction becomes more significant in addition to the hydrogen-bonding interaction. This leads to the enhanced packing of the alkyl side chains.

The variation of spherulite size seen here with the length of the alkyl side chain is very different from the trend that was observed with the simple carbamates studied before (Figure 2 of ref 5). In the latter, the spherulite size varied very little with the side chain length up to C12, and the increase was significant for C18. Note that in that case the (single) hydrogen-bonding motif was asymmetrically flanked by the alkyl chains until the length reached C18. However, in the present case, the alkyl chains are symmetrically positioned on either side of the double hydrogen-bonding motif. Thus, the alkyl chain dispositions contribute to the differences in the spherulitic morphology seen between the two cases. Spherulitic Growth Rates. To further investigate the influence of the alkyl side chains on the spherulitic crystallization, isothermal kinetic studies were performed using the Linkam hot stage. As an example, Figure 8 shows a series of micrographs recorded at different times for C12. This sample showed a Tm of 115 °C and a crystallization temperature of 102.8 °C in the DSC. For the isothermal crystallization kinetic study, the sample was heated to 135 °C, the temperature was reduced to 102.5 °C, and the micrographs were recorded at different times to follow the spherulite growth. It is seen that after 1 s the fibrillar growth is dumb-bell shaped. Extensive branching of the fibrils occurs with time. It appears that the entire growth is dominated by sequential branching of the lamellae. Similar experiments were performed with the other biscarbamates by crystallizing the samples about 0.5 °C below the Tc observed in the DSC. The variation of the size of the spherulites with time is shown in Figure 9a for the biscarbamates with C6, C8, C12, C16, and C18 side chains. It is seen that the growth is complete (due to impinging) within 10 s in the case of C8, whereas it takes about 29 s for C16 and C18 to complete the growth. It is also seen in this figure that the plots for C8 and C12 do not extrapolate to a spherulite size of zero at time ) 0 s. This is due to the extremely fast growth of the spherulite in less than 1 s. The growth rate plotted in Figure 9b shows that the maximum growth rate occurs with C8 and decreases significantly with the longer side chains. This accords with the maximum spherulite size seen with C8 in the case of quenched and slow-cooled samples (Figure 6). The significant contribution from the van der Waals interaction between the alkyl side chains leads to such dependence on the side chain length. Rate of Crystallization. In addition to the rate of growth of the spherulites during isothermal crystallization, the rate of crystallization is also influenced by the length of the alkyl side chain. Figure 10 shows the fractional heat of crystallization (R) as a function of time for C8 and C12 as the sample was held at the crystallization temperature in the DSC. As noted with the spherulite growth, the rate of crystallization is faster with C8 than C12. The t1/2 values are 0.13 and 0.66 min for C8 and C12, respectively. Orientation and a Model for Growth. It was noted before that the X-ray diffractometer traces of slow-cooled samples indicated orientation. The fibrillar growth of the spherulites

TABLE 4: IR Spectral Results for the N-H and CdO Groups for the Biscarbamates with Different Alkyl Chain Lengths H-bonded N-H sample description

peak position (cm-1)

C4 C6 C8 C12 C16 C18

3324 3322 3321 3321 3319 3319

H-bonded CdO

intensity

peak width at half-maximum

peak position (cm-1)

intensity

peak width at half-maximum

59.3 60.0 66.3 45.2 39.2 35.5

92.8 102.1 74.6 73.8 61.3 62.2

1682 1681 1682 1682 1682 1682

52.8 52.7 61.3 49.2 46.3 40.0

70.7 68.6 67.6 60.6 60.1 54.7

15258 J. Phys. Chem. B, Vol. 110, No. 31, 2006

Khanna et al.

Figure 8. Optical micrographs recorded during the isothermal crystallization of the C12 biscarbamates with time. Times (s) are shown in the figures.

shown in Figure 8 confirms this observation. In some cases, due to their large size, the spherulites could be seen with the naked eye. To gain some insight into the growth pattern of the spherulites, X-ray diffraction patterns were recorded on film using the flat-film camera. To this end, the sample was positioned on the beam collimator using the optical microscope so as to locate the fibrillar part in the path of the X-ray beam. Diffraction patterns of this type have been reported only in a few publications30 using microdiffraction or other sophisticated means. Here we used a simple flat-film camera. Figure 11a and 11b show the oriented X-ray patterns obtained in this manner for C6 and C12. The part of the spherulite that was in the X-ray beam is also shown. The d spacings of the reflections are marked. It is seen that in both cases the 4.6Å reflection is on the equatorial direction and the 3.8 Å reflection appears on the meridian. A schematic of these reflections with respect to the packing of C12 molecules is illustrated in Figure 4. The radial direction of the spherulite corresponds to the growth direction of the hydrogen-bonding plane, as evidenced by the 4.6 Å equatorial reflection. In addition to the reflection at 3.8 Å, those with larger spacing (e.g., 11.5 Å for C6 and 7.9 Å for C12) are also seen along the meridian. As the schematic in Figure 4 shows, the axis of the individual molecule (i.e., the direction of the alkyl chain) is along the tangential direction of the spherulites, indicating the origin of negative birefringence. Conclusions The results presented here demonstrate that single and multiple hydrogen-bonding motifs lead to different morphologies and that the symmetry and length of the side chains play an

Figure 9. (a) Growth of spherulitic radii with time for different biscarbamate samples. (b) Rate of growth of spherulite radii versus number of carbon atoms in the alkyl side chain.

Self-Assembling Carbamates with Alkyl Side Chains

Figure 10. Fractional heat of fusion versus time for C8 and C12 samples.

important role. We believe that studies of this type would provide insight into the factors that govern the morphology and enable control of the morphology for the functional context. We showed that in the case of simple carbamates blending of the homologous pairs led to reduction of the spherulite size, as required for application such as ink-jet printing. This could aid in the formulation of inks for such applications. The different morphological behavior of the biscarbamates would need different approaches for ink formulations. In the case of simple carbamates, the size of the spherulites does not increase significantly until the alkyl chain reaches a length that would impart symmetry on either side of the

J. Phys. Chem. B, Vol. 110, No. 31, 2006 15259 hydrogen-bonding group. With the biscarbamates studied here, such symmetry is inherent in the molecule and a gradual increase in spherulite size is seen. However, even if the disposition of the alkyl side chains was symmetric, the size of the spherulites, both with quenching and slow cooling, does not increase with alkyl chain length but shows a maximum with C8. We rationalize this behavior on the basis of the nucleation rate that is controlled by the relative contributions to the hydrogen-bond and van der Waals energy in the packing of two molecules. As the alkyl side chain length increases, the van der Waals attraction between these segments increases. This is supported by the shift of the CH2 frequencies to lower wavelengths with increasing length of the alkyl chain. As mentioned above, McKierman et al.26 noted that in the case of aliphatic polyurethanes, “dilution” of the hydrogen-bond contribution relative to van der Waals interactions occurs with an increase in the length of the alkyl segment. Although we are not dealing with a polymer in the present work but a small molecule model, a similar dilution or compensation occurs beyond an alkyl chain length of C8. Thus, we find a parallel between the crystallization and morphological behavior of these biscarbamates and aliphatic polyurethanes. Hence, this is a model system for such polymers. In the previous study on simple carbamates, kinetic measurements could not be made (with the facilities available to us) due to fast crystallization. The current study shows that the rates of both spherulite growth and crystallization depend on the side chain length, showing a maximum with an alkyl chain of C8. Due to the large size of the spherulites, we were able to record oriented X-ray diffraction patterns that show the hydrogen

Figure 11. X-ray diffractograms of (a) C6 and (b) C12 isothermally crystallized samples from the oriented fibrillar part (c) of the spherulite.

15260 J. Phys. Chem. B, Vol. 110, No. 31, 2006 bonding between the molecules to be along the radial direction of the spherulites. In this case, however, there was no difference in the direction of growth between the molecules with short and long alkyl chains (e.g., between C6 and C12). Acknowledgment. Financial support by the Natural Sciences and Engineering Research Council of Canada (NSERC) and Xerox Research Centre of Canada is gratefully acknowledged. M.M. was a recipient of an Ontario Government Scholarship for Science and Technology. Supporting Information Available: Brief description of the synthesis of the biscarbamates used in this study. This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) Lehn, J.-M. Supramolecular Chemistry; VCH: Germany, 1995. Ikeda, M.; Nobori, T.; Schmutz, M.; Lehn, J.-M. Chem. Eur. J. 2005, 11, 662. (2) Beijer, F. H.; Kooijman, H.; Spek, A. L.; Sijbesma, R. P.; Meijer, E. W. Angew. Chem., Int. Ed. 1998, 37, 75. Lehn, J.-M. Makromol. Chem. Macromol. Symp. 1993, 69, 1. (3) Llange, R. F. M.; van Gurp, M.; Meijer, E. W. J. Polym. Chem., Part A Polym. Chem. 1999, 37, 3657. (4) Bazuin, C. G.; Brandys, F. A.; Eve, T. M.; Plante, M. Macromol. Symp. 1994, 84, 183. (5) Moniruzzaman, M.; Goodbrand, B.; Sundararajan, P. R. J. Phys. Chem. B 2003, 107, 8416. (6) Moniruzzaman, M.; Sundararajan, P. R. Langmuir 2005, 21, 3802. (7) Gaylord, N. G. J. Org. Chem. 1955, 20, 546. (8) Martinek, T. W.; Klass, D. L. U.S. Patent 3,335,139, 1967. (9) Chambers, J.; Reese, C. B. Br. Polym. J. 1977, 9, 41. (10) Steichele, K. Belg. Patent 882,922, 1980; Chem. Abstr. 1980, 94, 66768d. (11) Saka, K.; Noda, K. Jpn. Kokai Tokkyo Koho JP 62 179 584, 1987; Chem. Abstr. 1987, 108, 39820r.

Khanna et al. (12) Tanaka, K.; Kano, Y.; Yoshida, K. Jpn. Kokai Tokkyo Koho JP 63 248 894, 1988; Chem. Abstr. 1988, 110, 79137w. (13) Kinoshita, H.; Sekiya, M.; Mishima, M. Eur. Pat. EP 274 756, 1988; Chem. Abstr. 1988, 109, 233979k. (14) Rohr, W.; Franke, A.; Giertz, H.; Amann, A. Brit Patent 1,396,523, 1975. (15) D’Silva, T. D. J. U.S. Patent 4,400,389, 1983. (16) Atwell, G. J.; Fan, J.-Y.; Tan, K.; Denny, W. A. J. Med. Chem. 1998, 41, 4744. (17) Chueca, A.; Baron, M.; Lopez, G. J. Photosynthesis Proceedings International Congress, 5th; Akoyunoglou, G., Ed.; Balaban Int. Sci. Serv.: Philadelphia, Pennsylvania; 1981; p 555; Chem Abstr. 1981, 97, 50917. (18) Furer, V. L. J. Mol Struct. 1998, 449, 53. Furer, V. L. J. Mol Struct. 1999, 513, 1. Furer, V. L. J. Mol Struct. 2000, 520, 117. (19) Matzger, A. J.; Kim, K. Polym. Mater. Sci. Eng. 2003, 89, 825. (20) Jpn. Kokai Tokyo Koho JP 58 201 758, 1983; Chem Abstr. 1983, 100, 174289z. (21) Tanaka, T.; Yoshitomi, T.; Hanada Y.; Ohashi, M.; Takeda, Y. Jpn. Kokai Tokkyo Koho JP 62 090 289, 1987; Chem Abstr. 1987, 107, 208963. (22) Goodbrand, B.; Boils, D.; Sundararajan, P. R.; Wong, R.; Malhotra, S. U.S. Patent 6,187,082, 2001. (23) Moniruzzaman, M.; Sundararajan, P. R. J. Phys. Chem. B 2005, 109, 1192. (24) Goodbrand, B.; Boils, D.; Sundararajan, P. R.; Wong, R. U.S. Patent 6,414,051, 2002. (25) Klug, H. P.; Alexander, L. E. X-ray Diffraction Procedures, John Wiley & Sons: New York, 1954. (26) McKierman, R. L.; Heintz, A. M.; Hsu, S. L.; Gido, S. P.; Penelle, J. Polym. Mater. Sci. Eng. 2001, 84, 416. (27) Alimova, L. L.; Atovmyan, E. G.; Filipenko, O. S. Kristallografiya 1987, 32, 97. (28) Coleman, M. M.; Lee, K. H.; Skrovanek, D. J.; Painter, P. C. Macromolecules 1986, 19, 2149. (29) Suzuki, M.; Nakajima, Y.; Yumoto, M.; Kimura, M.; Shirai, H.; Hanabusa, K. Langmuir 2003, 19, 8622. (30) Ishikawa, S.; Kurita, T.; Suzuki, E. J. Polym. Sci., Part B 1963, 1, 127. Tanaka, T.; Fujita, M.; Takeuchi, A.; Suzuki, Y.; Uesugi, K.; Doi, Y.; Iwata, T. Polymer 2005, 46, 5673.