Multiple Detection in Size-Exclusion Chromatography - American

Chapter 6

Downloaded by PENNSYLVANIA STATE UNIV on September 7, 2012 | http://pubs.acs.org Publication Date: November 4, 2004 | doi: 10.1021/bk-2005-0893.ch006

Application of Size-Exclusion Chromatography with Fluorescence Detection to the Study of Polymer Reaction Kinetics 1

2

1,

Ashok J. Maliakal , Ben O'Shaughnessy , and Nicholas J. Turro * 2

Departments of 'Chemistry and Chemical Engineering, Columbia University, New York, NY 10027

SEC with fluorescence detection has been used to study the kinetics of polymer chain end reactions in solution. The presence of afluorescentpyrene label in a styrene terminated polystyrene allows for sensitive detection of both starting material and product, which are resolved as a function of molecular weight through SEC. Hence measurements can be performed under pseudo-first order conditions allowing a simple determination of bimolecular polymer end-end reaction rate constants. This method is applicable to measurement of both activation controlled and diffusion controlled polymer reaction kinetics (i.e. radical-radical reactions).

The measurement of polymer end-end reaction rates has been of interest to experimental(1,2) and theoretical polymer scientists/3-5) Fundamental theories have been developed to predict the effect of chain length on reaction rates/4, However, several experimental difficulties have hindered the measurement of polymer chain end reaction kinetics (e.g., synthesis of suitable monodisperse end-labeled samples and low sensitivity of analytical techniques for the detection of polymer end groups). Although photophysical approaches have been used to measure interpolymer chain end reaction rates/2,7-/0), these methods are

114

© 2005 American Chemical Society

In Multiple Detection in Size-Exclusion Chromatography; Striegel, A.; ACS Symposium Series; American Chemical Society: Washington, DC, 2004.


115 limited in dynamic range by thefluorescenceand phosphorescence lifetimes of the probes. In order to address the issues of sensitivity and dynamic range, an SEC-based method has been developed which resolves reactant and product based on molecular weight and which employs fluorescence detection in the analysis of the disappearance of starting material and the formation of the end to end coupling product/7/, 12) Previous studies of the kinetics of activation controlled polymer end-end reactions observed either loss of starting material//5,) or product formation//^) but were unable to simultaneously measure both. The current chapter discusses the application of SEC with fluorescence detection (SEC-F) to the study of both activation and diffusion controlled polymer chain end reaction kinetics. The chain length dependence has been determined for the activation controlled reaction of polystyryl lithium 1 [PSLi] (see Figure 1 ; Ν = degree of polymerization) with a second polystyrene molecule 2 (referred to as PSPYSTY) which is selectively labeled with one pyrene fluorophore and contains a reactive styrene functionality at the chain end///,)

S B P -Li SN

•

C

J

- WT v

1 Degree of Polymerizations Ν

IVM Pyrene

PSPYSTY 2 degree of polymerization = M

K , M

3 Degree of Polymerization M+N

Figure 1. Synthesis ofStyrene Endlabeled-Pyrene-Labeled Polystyrene, PSPYSTY(2). The high sensitivity of the pyrene probe 2 permits measurement of pseudofirst order rate constants for polymer end-end reactions, allowing for relatively simple kinetic analysis. Surprisingly this study reveals an increase in the bimolecular reaction rate constant with increasing molecular weight. This result has been discussed in the context of the complex supramolecular environment present in the PSLi/benzene system///,) In switching to a fast reacting system (approaching diffusion controlled), mixing of reagents to start the reaction, and time resolution become important issues. Time-resolved detection methods exist (i.e., time resolved IR//5) time resolved ESR, (16) and laser flash photolysis followed by timeresolved UV-Vis spectroscopy(7 7) and photophysical quenching techniques.)(79,18-20)) However these techniques again involve significant signal to noise issues and are limited by the excited state lifetime of the probe employed. In the case of photophysical methods, relating quenching rate constants to reaction rate



116 constants is complicated by the fact that quenching rate constants occur at a distance due to dipole-dipole mechanisms// 7) where as chemical reaction requires close contact between chain ends to accomplish reaction//,) To address these issues, photoinitiators have been used to generate fluorescent labeled polymers containing a chain end radical (macroradicals), and the reactions of these macroradicals have been monitored using SEC-F. The dynamic range in this case is no longer an issue, since fluorescence is used only as a label whose transfer from starting material to product is being monitored after completion of reaction. To address the issue of time resolution, a method of competitive kinetics is proposed for extracting the rate dependence from these fast polymer chain end reactions. Although actual rate constants and chain length dependences have not yet been achieved, the viability of the SEC-F method in the case of fast reactions is demonstrated.

Experimental Materials. Unless specified, compounds were purchased from Aldrich. Styrene was dried over calcium hydride, distilled onto dibutylmagnesium, and then distilled under reduced pressurefreshprior to use. Benzene was dried over calcium hydride, distilled over PSLi, and distilled fresh prior to use. THF (Acros) was dried over potassium hydride. l-(l-Phenyl-vinyl)-pyrene (4) was synthesized using a modified literature procedure.(21). Synthesis of Tosylate 6 is described elsewhere/22) The synthesis of PSPYSTY (number averaged molecular weight M =2300, degree of polymerization M = 22) has been described previously//1) N

PSU1

4

»

PSPYPI 7

Figure 2. Synthesis of PSPYPI 4. Synthesis of PSPYPI (7). l-(l-Phenyl-vinyl)-pyrene 4 was freeze-dried from dry degassed benzene under vacuum. After this 10 ml of dry degassed THF is added. In a separate flask, the tosylate 6 wasfreeze-driedfromdry degassed benzene as well; 2 mL of dry degassed THF is added to this flask. PSLi was synthesized by addition of 0.65 mmol of s-butyl lithium to 1.9 mL of



117 styrene (16.25 mmol) in 5 mL of benzene which is cooled in an ice bath. Styrene and benzene were both dried over calcium hydride andfreshlyvacuum distilled prior to use. The appropriate amounts of each reagent are transferred via syringe into the flame dried reaction flask. The reaction is allowed to warm to room temperature an aged for 2 hours (reddish orange solution due to PSLi anion). 2 mL of the resulting orange solution is transferred via flame-dried syringe to the flask containing a THF solution of 4. A dark blueish purple solution is obtained instantly, this solution is aged 10 mins and then transferred via cannula to the solution of 8 in THF held at -78°C using a dry ice acetone bath. The blueishpurple solution (presumable the di-aryl-anion 5) decolorizes rapidly. The reaction is monitored by TLC, and 6 is observed to disappear. PSPYPI is recovered as a pale yellow solid after several precipitationsfrommethanol. SEC indicates one peak with polydispersity 1.2, M = 2000. UV-Vis and fluorescence spectra confirm incorporation of pyrene into polymer. UV-Vis and excitation spectra do indicate the presence of a highly colored impurity. ^ - N M R is consistent with the product. The benzylic proton resonances are split from the expected singlet due to diastereotopic interactions from the atactic polystyrene backbone known to extend over up to 3 styrene residues. (23)(24) N

Instrumentation. Size exclusion chromatography was performed on a Polymer Labs SEC with a GTI/Spectrovision FD-500 fluorescence detector. 2 PL mixgel 5_ C and 1 PL gel 5_ 100A columns were used in series, and calibration was performed with polystyrene standards (Polymer Labs). Measurement of the Rate of Reaction of 1 with 4. Polystyryl lithium 1 is generated by reaction of an appropriate amount of s-butyllithium with 5.85 mL (50 mmol) of styrene in 10 mL of dry degassed benzene. SEC molecular weight (M ) of resultant PSLi is used to confirm the concentration of s-butyl lithium added. After reaction is complete (~2 hrs), this solution is maintained at constant temperature of 30 ± 0.2°C using an IKA ETS-D4 temperature controller. In a separate flask, PSPYSTY (2) (5 mg, 2.2x10" mmol) is dissolved in a small amount of dry degassed benzene, and added to the solution of 1 in benzene. The addition time is marked with a stopwatch, and aliquots are removed subsequently and quenched onto methanol. A small degree of coupling reaction (few percent) is observed due to the formation of free radical chains ends resulting from oxidation of anionic chains, since the methanol employed is not rigorously degassed (see shoulder in Figure 3 for product peak 3).(25) However, this is not expected to affect the kinetic analysis. The samples thus generated are dissolved in THF and filtered through a short plug of silica gel to remove lithium salts prior to injection into the SEC. N

3


118 Reaction of PSPYSTY (2) and (diphenyl-phosphinoyl)-(2 4 6-trimethylphenyl)-methanone (15) in Rayonet Reactor 9

9

PSPYSTY (2) was dissolved in benzene (spectroscopic grade Aldrich). The O.D. for solution is 1.78 @313 nm, which translates to \.2x\0~ M PSPYSTY based on pyrene absorption (8p = 1.5xl0 M in nonpolar solvent)(26) 8 was added (O.D. 0.07 at 381 nm) and mixture is degassed by argon bubbling, and irradiated for 3 mins at 420 nm in Rayonet Carousel Photoreactor with 420 nm lamps (Strontium Pyrophosphate/Europium lamps emission max at 420 nm, 34 nm width at 50% height, manufactured by Southern New England Ultraviolet, Branford CT) UV-VIS absorption indicates complete consumption of 8 and yellowing of the solution is observed. Reaction is continued by adding more 8 (0.07 Abs at 381 nm). After further irradiation (-3 mins), reaction is stopped and products injected onto SEC (see Figure 11). SEC fluorescence conditions (excite at 330 nm, emission at 450 nm). 4

4


y

Results and Discussion

Measurement of Activation Controlled Polymer Chain End Reaction Rates Previous efforts^/1) from our research group have measured the rate of polymer chain end reactions as a function of chain length using SEC-F for the activation controlled reaction of PSLi (degree of polymerization N) with a macromolecular fluorescent labeled styrene (PSPYSTY; 2). Reaction of 1 with fluorescently labeled PSPYSTY (2) is clean and yields only one product 3 after quenching with methanol (see Figure 1). Both PSPYSTY and 3 can be resolved by SEC as a function of molecular weight, and observed by fluorescence detection (excitation at 330 nm, emission at 400 nm) in a regime in which there is negligible background signalfrompolystyrene. The reaction illustrated in Figure 1 was employed for measurement of rate constants for the end to end polymer coupling//7,) PSLi solutions of varying chain lengths were heated to 30°C and kinetics were measured at this temperature. A solution of PSPYSTY (2) dissolved in a small amount of benzene was added to the PSLi solution. Aliquots were withdrawn periodically and quenched into methanol. The resulting samples were injected into the SEC to produce a time series of SEC traces (representative examples are shown in Figure 3). The fluorescence of the pyrene is observed in the starting material PSPYSTY (2) and product 3, but the PSLi is effectively non-emissive under


119


these detection conditions (excitation at 330 nm and observation at 400 nm). For this reason we are able to operate under pseudo-first order kinetic conditions with a 50-100 fold excess of PSLi as compared to PSPYSTY (2) without any significant interference in the SEC fluorescence analysis from the PSLi. The area integrations of peaks 2 and 3 in Figure 3 are proportional to the concentrations of starting material and product in the reaction. These areas are used to calculate thefractionconversion as a function of time (see Figure 4). The conversion data for the starting material and product are fit with exponential curves (see Figure 4).

Τ

I

I

I

16

18

20 Time (min)

22

Γ 24

Figure 3. SEC Traces with Fluorescence Detection (Excitation at 330 nm, Observation at 400 nm) at Several Reaction Times. Peak for Starting Material 2 Decreases with Reaction Time and Peak for Products 3 Increases with Reaction Time. Reaction Conditions: Benzene, 30°C (Adaptedfrom Reference 11.) (Copyright 2003 American Chemical Society.)

As the chain length was varied, the concentration of PSLi also changed, and for this reason it was important to assess the dependence of the rate constant on PSLi concentration. In order to confirm the PSLi concentration dependence (which has been reported of the order of 0.5 in polystyrene propagation^7-29) as well as varying between 0.48 to 0.87(30)), the dependence of ko on [PSLi] at a constant molecular weight ( M = 10 Κ; Ν « 100) was measured. Although there is much discussion in the literature regarding aggregation state and reaction mechanism, under our experimental conditions we observe a roughly linear bs

N


120


dependence of Ull)

ko

bs

0

versus [PSLi] which is consistent with the rate equation

10

20

30 Time (min)

40

50

60

Figure 4. Representative Kinetic Trace and First Order Exponential Fit for Decrease of Starting Material 2 and Growth of Product 3 as Measuredfrom Area of Peaks in Figure 1. (M for PS Li (1) = 4500 ± 100 amu, M for 3(M + N) = 6900 ± 200) (Adaptedfrom Reference 11.) (Copyright 2003 American Chemical Society.) N

N

N

This experimentally determined relationship was used to calculate the bimolecular rate constant of reaction k ,N as a function of chain length at constant volume fraction polystyrene. (In order to simplify units, the approximation of a first order dependence in [PSLi] was made in calculating k , N from kobs). These results are presented in Figure 5. It is interesting to note that the bimolecular rate constant k ,N increases as the degree of polymerization increases from Ν = 31 to 246 ( M = 3.3 K-25.6 K) from 3.6 M" min to 10.3 M" 'min' ± 1 M min . The dashed line indicates the transition between dilute and semi-dilute which occurs for polystyrene at this concentration in benzene at approximately 4.6 K. Previous studies have employed the use of SEC-F to explore coupling processes in high temperature melt processing (31-33) and at the thin film M

M

M

!

1

N

1

-I

,


121 interface/^ Our work has extended the SEC-F detection methodology to measure bimolecular rate constants directly for activation controlled polymerpolymer reactions in solution.

12-1

8


if

l

10

6-

Λ

"1

20-

ι

π

ιο M

N

15 (g/mol)

~l— 20

25x10

3

Figure 5. Chain Length Dependence of Bimolecular Polymer End-End Reaction Rate Constant k (Plotted with Estimated Error ± 1 M^in ). M of PSPYSTY 2 = 2300 ± 100, M = 22. M of PSLi (1) Variedfrom 4K-25K (N = 40-240). Dotted Line Indicates Transition Between Dilute and Semi-dilute Regimes (Adaptedfrom Reference 11.) (Copyright 2003 American Chemical Society.) 1

MtN

N

N

Several possible reasons were considered for the increase of k ,N with chain length///,) However, a change in the supramolecular structure of 1 with increasing chain length was most consistent with the experimental data. (30,3436) In this hypothesis, as Ν increases, the aggregation number of supramolecular micellar aggregates of 1 are expected to decrease due to increased excluded volume repulsive interactions of the polystyryl chains/55,5tf) There is a possibility that in these smaller micelles, 1 is more reactive (See Figure 6).(36) Attempts to investigate kinetics at longer chain lengths were limited by an increase of viscosity to the point where reaction was limited by stirring. For example, an experiment performed with M = 55K PSLi chains (30% by volume PSLi) resulted in a viscous solution where stirring was no longer efficient. PSPYSTY (2) did not mix homogeneously as evidenced by the observation of nonexponential and erratic kinetics for disappearance of starting material and appearance of product. At this chain length and concentration, the PSLi solution M

N


122 behaves like an entangled transient network, as predicted by polymer theory.(3) The stir rate (-100-800 min") is now faster than the relaxation of the polymer network, so in order for stirring to occur, the stir bar must either break the polymer chains or wait for entangled chains to relax. The dependence of fractional conversion on the location of sampling indicates incomplete mixing. Transient network formation presents a significant challenge to the study of polymer reactions in the entanglement region, a problem similar to the study of polymer-polymer reactions in the solid state.(2,37)


1

Figure 6. Effect of Chain Length on Polybutadienyl Lithium Micellar Aggregation State.

Despite the high activation energy for addition of PSLi to the styrene moiety, the reaction becomes diffusion controlled due to the slow polymer chain end diffusion under entanglement constraints. Approaches to studying diffusion controlled polymer reactions using SEC-F are discussed in the next section.

Approaches to Measuring Diffusion Controlled Polymer Reaction Kinetics Fast reaction processes can be measured through competitive kinetics. Concurrent reactions of mixtures have been used to assess the relative rate constants between a reactive intermediate and two or more different competitive reaction partners.(3$ The system envisioned for measuring the chain length effect for high Q rate constants is based on this concept of the kinetics of concurrent reactions of mixtures. A monodisperse fluorescent labeled photoinitiator end-labeled polymer (P-R) (degree of polymerization M) was sought which upon photolysis yields a macroradical Ρ- (see Figure 7). This macroradical is formed in the presence of a variable but high concentration of a reactive small molecule (S) as well as a variable but high concentration of an analogous polymeric reactive molecule (M) (degree of polymerization N). The simplified kinetic scheme is illustrated in Figure 7.


123 hv

P-R ρ· +

p.

P. + R*

•

s

k

°

» P-S

k

" » p.

M

+M


Figure 7. Concurrent Reaction of Ρ · with competitive small (S) and macromolecular (M) reagents.

After a reaction time t, measurement of the relative amounts of P-M (degree of polymerization M+N) and P-S (degree of polymerization M) would be performed using SEC with fluorescence analysis. The ratio of P-M to P-S could be related to the ratio of rate constants k /k through equation 2. The reaction of P- with itself or with R- could be minimized through high relative concentrations of S and M (similar to the pseudo-first order kinetic conditions employed in the previous section). N

0

k _ [PM][S] N

Κ

( 2 )

[PS][M]

Performing the experiment at various degrees of polymerization for P-R and M would permit assessment of the chain length dependence of this fast reaction. With this kinetic model in mind, two approaches were attempted to generate a suitable macroradicalΡ·. In the first approach, the monodisperse photoinitiator terminated pyrene labeled polystyrene PSPYPI (7) was synthesized. The commonly used α-alkoxy-ketone photoinitiator moiety was incorporated into the polymer 7 (see Figure %).(39,40)(41) Typically this photoinitiator, upon absorption of light, undergoes rapid intersystem crossing and Norrish Type I cleavage to form the macroradical-radical pair illustrated in Figure 8. However triplet energy transfer from the photoinitiator to the pyrene moiety prevented PSPYPI from cleaving, rendering the photoinitiator inactive/22) If the photoinitiator is separated from the chromophore, then it should be photoactive in the presence of a pyrene chromophore, as was found to be the case using an external photoinitiator.


124

MACRORAOICAL

Figure 8. Compound chosen as P-R for competitive kinetics to measure k^/k . Downloaded by PENNSYLVANIA STATE UNIV on September 7, 2012 | http://pubs.acs.org Publication Date: November 4, 2004 | doi: 10.1021/bk-2005-0893.ch006

Q

PS-PY-STY 2

Figure 9. External Photoinitiation to generate macroradical PSPYPI.

In a second strategy, the macroradical was generated through reaction of an external photoinitiator (I, see Figure 9) with PSPYSTY 2. In this strategy an external photoinitiator was photolysed independent of the polymeric species. The resulting radicals add rapidly to the styrene terminated polystyrene (PSPYSTY) to generate the macroradical illustrated in Figure 9. The fate of the macroradical 9 can be tracked with the aid of thefluorescentlabel using SEC analysis. There are several different types of radical photoinitiators.(¥2) The following considerations were of prime importance in our selection of an external photoinitiator. First, a long wavelength absorption was desirable which could be selectively excited independent of PSPYSTY. Secondly, we sought a radical which adds rapidly to styrene in order to rapidly generate the macroradical 9. These two criteria coupled with the desire for a photoinitiator with a high quantum yield lead us to choose (diphenyl-phosphinoyl)-(2,4,6trimethyl-phenyl)-methanone (8) as the external photoinitiator. Excitement of 8 can be achieved independent of PSPYSTY due to the long wavelength absorption band of 8 (at 420 nm). Furthermore literature references(^2) indicate that the resultant phosphinoyl radicals (see Figure 10) add to the styrene moiety at a rate of ~10 M ' V . Thus external initiation is 7

1


125 another route to generating monodisperse macroradicals rapidly. The resultant macroradicals (10) are capable of dimerizing (or undergoing disproportionation) as may be observed from SEC analysis of reaction (see Figure 11). After exhaustive photolysis in THF, an aliquot of the reaction mixture is diluted and injected into the SEC. The early peak in Figure 11 (retention time 20.7 minutes) corresponds to a molecular weight ( M = 5000 ± 100) consistent with the dimeric product 11 in Figure 10» The latter peak ( retention time 21.9 minutes) corresponds to the molecular weight of the starting material 2 ( 2300 ± 100 amu). This peak most likely corresponds to the product resultingfromthe disproportionation reaction of the macroradical 10 and/or reaction of 10 with 2,4,6-trimethylbenzoyl radical (see Figure 10). Although the dimerization reaction of 10 shows in concept the potential of using fluorescence SEC to monitor kinetics of polymer chain end reactions in the fast reaction regime, these particular reaction conditions involve too many competing reactions to allow for the simple extraction of k the chain length dependent interpolymer end reaction rate constant. Our lab is currently seeking suitable compounds for the roles of S and M in Figure 7. The nitroxides which have been employed previously as radical traps are promising candidates (see Figure \2).(43,44)


N

N

PSPYSTY DIMER11

Figure 10. Reaction Pathway for Photolysis of 8 in the presence of PSPYSTY (2).


126

PSPYSTY 2 \ ^before photolysis

PSPYSTY DIMER 11 after photolysis


—Γ" 18

I 20

\

r~ 22

24

Retention Time (mins)

τ 26

Figure 11. SEC Analysis of Photolysis of 8 in the Presence of PSPYSTY 2 (see Figure 10 for proposed reaction pathways).

Figure 12. Plan for Using Nitroxide Based Radical Traps (R' = H (TEMPO), or polystyrene, R ' = PS) to Trap Macroradical.

Summary and Conclusions SEC based measurement of kinetics using fluorescence detection is demonstrated to be a powerful method to extract bimolecular rate constants of polymer end-end reactions. The sensitivity of fluorescence detection permits kinetic study under pseudo-first order conditions, allowing for a simple kinetic analysis. This method has been applied to the problem of measuring the chain length dependence of polymer end-end reaction rates. Attempts to measure kinetics in the entanglement regime were thwarted by the inability to mix starting material and product efficiently to create



127 homogenous reaction. Nonexponential kinetics result due to heterogeneity in reaction environment, a situation similar to that observed for polymer reactions in the solid state/2,57; For fast reactions, photoinitiation gets around the problem of mixing dependent reaction rates by premixing reagents and homogeneously initiating the reaction using light. A concurrent competing reaction scheme was proposed which would allow for determination of the chain length dependence of inter polymer chain end kinetics. An initial approach to Ρ-Λ involving the attachment of a photoinitiator in close proximity to a pyrene end-labeled polymer failed due to quenching of the photoinitiator by the pyrene. Subsequent attempts employing an external photoinitiator allowed for the formation of fluorescent labeled macroradical as observed by SEC-F. Although the generation and subsequent reactions of PSmacroradicals 11 could potentially be studied kinetically, the requirement of a large amount of external photinitiator 8 complicated the kinetic scheme. The current work demonstrates that fast reactions can be monitored by SEC-F. In order to make these systems more transparent to kinetic analysis, we are currently altering the distance between photinitiator and fluorescent label in order to minimize triplet energy transfer quenching in the photoinitiator. A second possible solution to get around utilizing external photoinitiators is to use a fluorophore with a triplet energy higher than the photoinitiator. This would make triplet energy transfer thermodynamically unfavorable and permit cleavage of the photoinitiator triplet. Furthermore efforts are underway to synthesize suitable macromolecular radical traps (M) in order to be able to measure fast interpolymer chain end reaction rates using the concurrent reaction scheme illustrated in Figure 7.

Acknowledgements The authors thank the National Science Foundation (Grants CHE-00-91460 and CHE-'01-10655) for its support of this research. AJM thanks the NSF for a graduate fellowship.

References (1) de Kock, J. B. L.; Van Herk, A. M.; German, A. L. J. Macromol. Sci.Polym. Rev 2001, C41, 199-252. (2) Mita, I.; Horie, K. J. Macromol. Sci., Rev. Macromol. Chem. Phys. 1987, C27, 91-169.



28 (3) de-Gennes, P. Scaling Concepts in Polymer Physics; Cornell University Press: Ithaca, NY, 1985. (4) Friedman, B.; O'Shaughnessy, B. Int. J. Mod. Phys. B. 1994, 8, 2555-2591. (5) Khokhlov, A. Makromol. Chem. Rapid. Commun. 1981, 2, 633. (6) Wang, Y. C ; Morawetz, H. Macromolecules 1990, 23, 1753-1760. (7) Winnik, M.; Sinclair, Α.; Beinert, G. Macromolecules 1985,18, 1517-1518. (8) Sinclair, Α.; Winnik, M.; Beinert, G. J. Am. Chem. Soc. 1985, 107, 57985800. (9) Gebert, M.; Torkelson, J. Polymer 1990,31,2402-2410. (ΙΟ) Yu, D.; Torkelson, J. Macromolecules 1988, 21, 852-853. (11) Maliakal, Α.; Greenaway, H.; O'Shaughnessy, B.; Turro, N. J. Macromolecules 2003, 36, 6075 - 6080. (12) Moon, B.; Hoye, T.; Macosko, C. J. Polym. Sci. Part. A. Polymer Chemistry 2000, 38, 2177-2185. (13) Okamoto, Α.; Shimanuki, Y.; Mita, I. Eur. Polym. J. 1982, 18, 545-548. (14) Okamoto, Α.; Toyoshima, K.; Mita, I. Eur. Polym. J. 1983, 19, 341-346. (15) Sluggett, G.; Turro, C.; George, M.; Koptyug, I.; Turro, N. J. Am. Chem. Soc. 1995,117,5148-5153. (16) Liu, Z.; Weber, M.; Turro, N.J.; O'Shaughnessy, B. in "Photoinitiated Polymerization", eds. K.D. Belfield and J.V. Crivello, ACS Symposium Series 847, Washington, DC 2003. (17) Turro, N. J. Modern Molecular Photochemistry; University Science Books: Sausalito, 1991. (18) Horie, K.; Mita, I. Macromolecules 1978, 11, 1175-1179. (19) Mita, I.; Horie, K.; Takeda, M. Macromolecules 1981, 14, 1428-1433. (20) Gebert, M. S.; Yu, D. H. S.; Torkelson, J. M. Macromolecules 1992, 25, 4160-4166. (21) Quirk, R.; Schock, L. Macromolecules 1991, 24, 1237-1241. (22) Maliakal, A. Ph. D. Thesis, Dept. of Chemistry; Columbia University: NY, 2003. (23) Moad, G. Chem. Aust. 1991, 58, 122-126. (24) Bevington, J.; Lyons, R.; Senogles, E. Eur, Polym. J. 1992, 28, 283-286. (25) Fetters, L.; Firer, E. Polymer 1977, 18, 306-307. (26) Murov, S.; Carmichael, I.; Hug., G. Handbook of Photochemistry; M . Dekker: New York, 1993. (27) Worsfold, D.; Bywater, S. Can. J. Chem. 1960, 38, 1891-1900. (28) Morton, M.; Fetters, L. J. Polym. Sci. Part. A. Polymer Chemistry 1964, 2, 3311-3326. (29) Morton, M.; Fetters, L.; Pett, R.; Meier, J. Macromolecules 1970, 5, 327332. (30) Fetters, L. J.; Huang, J. S.; Stellbrink, J.; Willner, L.; Richter, D. Macromol. Symp. 1997, 121, 1-26.



129 (31) Gray, M.; Kinsinger, M.; Torkelson, J. Macromolecules 2002, 55, 82618264. (32) Yin, Z.; Koulic, C.; Jeon, H.; Pagnoulle, C.; Macosko, C.; Jerome, R. Macromolecules 2002, 55, 8917-8919. (33) Schulze, J.; Moon, B.; Lodge, T.; Macosko, C. Macromolecules 2001, 34, 200-205. (34)Stellbrink, J.; Willner, L.; Jucknischke, O.; Richter, D.; Lindner, P.; Fetters, L. J.; Huang, J. S. Macromolecules 1998, 31, 4189-4197. (35)Stellbrink, J.; Willner, L.; Richter, D.; Lindner, P.; Fetters, L. J.; Huang, J. S. Macromolecules 1999, 52, 5321-5329. (36)Stellbrink, J.; Allgaier, J.; Willner, L.; Richter, D.; Slawecki, T.; Fetters, L. J. Polymer 43, 7101-7109. (37) Horie, K.; Mita, I. Adv. Polym. Sci. 1989, 88, 77-128. (38) Espenson, J. Chemical Kinetic and Reaction Mechanisms; McGraw Hill: New York, 1981. (39) Fouassier, J. Euro. Coatings Journal 1996, 723-726. (40)Turro, N.; Wu, C. J. Am. Chem. Soc. 1995, 117, 11031-11032. (41) Wu, C. Ph. D. Thesis, Dept. of Chemistry; Columbia University: New York City, 1994. (42) Reetz, I.; Yagci, Y.; Mishra, M. Handbook of Radical Vinyl Polymerization; Marcel Dekker, Inc.: New York, 1998; Vol. 48. (43) Braslau, R.; Anderson, M.; Rivera, F.; Jimenez, Α.; Haddad, T.; Axon, J. Tetrahedron 2002, 58, 5513-5523. (44)Moad, G.; Solomon, D. Chemistry of Free Radical Polymerization; Pergamon: Oxford, U.K., 1995.


Multiple Detection in Size-Exclusion Chromatography - American

Recommend Documents