Direct Quantification of Molar Masses of Copolymers by Online Liquid

Dec 17, 2013 - Fax: +492317553771. ... Online LCCC-NMR and SEC-NMR are compared regarding the determination of molar masses of block copolymers...
0 downloads 0 Views 1MB Size
Download PDF
Article pubs.acs.org/ac

Direct Quantification of Molar Masses of Copolymers by Online Liquid Chromatography under Critical Conditions−Nuclear Magnetic Resonance and Size Exclusion Chromatography−Nuclear Magnetic Resonance Mathias Hehn,† Thomas Wagner,‡ and Wolf Hiller*,† †

TU Dortmund, Department of Chemistry and Chemical Biology, Otto-Hahn-Straße 6, 44227 Dortmund, Germany Max-Planck-Institute for Polymer Research, Ackermannweg 10, 55128 Mainz, Germany

‡

S Supporting Information *

ABSTRACT: Online LCCC-NMR and SEC-NMR are compared regarding the determination of molar masses of block copolymers. Two different direct referencing methods are particularly demonstrated in LCCC-NMR for a detailed characterization of diblock copolymers and their co-monomers. First, an intramolecular reference group was used for the direct determination of block lengths and molar masses. For the first time, it was shown that LCCC-NMR can be used for an accurate determination of Mw and Mn of copolymers. These data were in perfect agreement with SEC-NMR measurements using the same intramolecular referencing method. In contrast, the determination of molar masses with common relative methods based on calibrations with homopolymers delivered inaccurate results for all investigated diblock copolymers due to different hydrodynamic volumes of the diblock copolymer compared to their homopolymers. The intramolecular referencing method provided detailed insights in the co-monomer behavior during the chromatographic separation of LCCC. Especially, accurate chain lengths and chemical compositions of the “invisible” and “visible” blocks were quantified during the elution under critical conditions and provided new aspects to the concept of critical conditions. Second, an external reference NMR signal was used to directly determine concentrations and molar masses of the block copolymers from the chromatographic elution profile. Consequently, the intensity axes of the resulting chromatograms were converted to molar amounts and masses, allowing for determination of the amount of polymer chains with respect to elution volume, the evaluation of the limiting magnitude of concentration for LCCC-NMR, and determination of the molar masses of copolymers.

O

case, a molar mass determination can be performed by using a calibration with defined homopolymers (which corresponds to the relative procedure in SEC). Several publications demonstrate molar mass determinations of block copolymers by LCCC and compare the results to SEC. Pasch et al. obtained the same molar masses for LCCC and SEC by studying poly(decyl methacrylate)-b-poly(methyl methacrylate)s (PDMA-b-PMMA) and polystyrene-b-poly(methyl methacrylate)s (PS-b-PMMA).7−9 Falkenhagen et al. also found identical results with both methods for poly(methyl methacrylate)-b-poly(tert-butyl methacrylate)s (PMMA-bPtBMA).10 Differences for the molar masses determined by LCCC and SEC were reported by Chang et al.11−13 and also by the authors.3 In these cases, it was seen that the co-monomer under critical conditions influenced the other block, which resulted in smaller molar masses due to differences of the chromatographic distribution coefficients of the block in SEC

ne essential aim in polymer characterization is the determination of the average molar masses. Size exclusion chromatography (SEC) is the most common used technique in this context. The SEC system is calibrated with polymer standards and is therefore a relative method. In the case of diblock copolymers, SEC will provide molar mass information for the whole copolymer when either multiple detection (e.g., ultraviolet (UV) and refractive index (RI))1 or NMR detection2−4 is applied for individually detecting the different block co-monomers. Another approach used light scattering for the molar mass determination.5 On the other hand, liquid chromatography under critical conditions (LCCC) is suitable to study only a single block of a copolymer. In this case, one block is in critical mode which is defined as chromatographically invisible. This mode is particularly characterized by a molar mass independence of this block. The critical conditions are experimentally established by identical elution times of homopolymers of different molar masses with the same structure as the critical block. The other visible block, however, is either in liquid adsorption chromatography (LAC) or in the SEC regime.6,7 In the latter © 2013 American Chemical Society

Received: July 31, 2013 Accepted: November 30, 2013 Published: December 17, 2013 490

dx.doi.org/10.1021/ac402396r | Anal. Chem. 2014, 86, 490−497

Analytical Chemistry

Article

mode, KSEC,block, and the corresponding homopolymer, KSEC,homopolymer.14 Consequently, the use of a calibration curve derived from homopolymers can lead to inaccurate molar masses of the co-monomers. Therefore, other methods of molar mass determination have to be taken into account. One possibility is the NMR detection of a unique reference group (for example, an end group) as an intramolecular reference for direct chain length calculations. This direct approach was first demonstrated by Hatada et al. in SEC-NMR investigations on PMMA homopolymers.15−17 Similarly, Hiller et al. determined the chain length distribution of small poly(ethylene oxide)s by LAC-NMR and SEC-NMR.18 Many different LC-NMR applications of polymers are summarized in a review.19 Several papers published by the authors also described the relative molar mass determinations by LCCC-NMR.20−24 They demonstrated the block length calculations of the diblock copolymers PS-b-PMMA,20,21 polyisoprene-b-poly(methyl methacrylate)s (PI-b-PMMA),22 as well as polystyrene-bpolyisoprenes (PS-b-PI).24 LCCC-NMR is particularly useful for the characterization of block copolymers because it allows for the separation of the copolymer and the contaminating precursor. Up to now, however, the direct referencing method has been used for neither chain length calculations in LCCCNMR nor the characterization of block copolymers in general. Further NMR referencing methods can be applied for the determination of concentration and mass in high resolution NMR. For example, weighed internal references can be used. However, these compounds must be soluble in the same solvent, they should be chemically inert, and their NMR signals must not overlap with signals of the sample.25 Due to these conditions, the use of reference substances is restricted. As an alternative, an Electronic REference To access In vivo Concentrations (ERETIC) was introduced by Akoka et al.26,27 In this case, an artificial NMR signal is created electronically, which is calibrated against a substance of known quantity. This signal can be easily adjusted according to frequency, intensity, phase, and width and was used in various studies, e.g., metabolites,26,28−32 reaction kinetics,33,34 and hydrogen isotope determination.35−38 Furthermore, ERETIC was implemented in 2D NMR,39 diffusion ordered NMR spectroscopy (DOSY),40,41 solid state NMR,42,43 and magnetic resonance imaging (MRI).44 In flow NMR, an application was presented concerning the investigation of reaction kinetics.34 However, to the best of our knowledge, ERETIC has never been applied for the quantification of polymers in online LC-NMR. In this publication, molar masses of diblock copolymers are determined by LCCC-NMR and SEC-NMR using both the common relative and the direct referencing method. In particular, the determination of Mw and Mn will be compared for both methods. In addition, ERETIC is implemented in LCCC-NMR measurements to obtain molar amount and mass related chromatograms. The methods are demonstrated for low molar mass diblock PI-b-PMMA copolymers of different PMMA block lengths containing a single diphenylethylene (DPE) unit per copolymer chain for the direct method of intramolecular referencing.

Figure 1. 1H NMR spectra of the PI-b-PMMA samples 1 (a) and 4 (b) with the assignment according to the structure of the copolymer.

molar mass of the PI block of 2 kg·mol−1 (determined by SEC) and varying molar masses of the PMMA block. Table 1 contains the chemical compositions, block lengths, and number average molar masses of all samples obtained by static 1H NMR. Figure 1 shows the NMR spectra of samples 1 and 4 recorded at 500.13 MHz in CD2Cl2. In addition to the block copolymers, a PI homopolymer endcapped with DPE was produced (sample 5). LCCC-NMR and SEC-NMR. LCCC-NMR and SEC-NMR experiments were performed with an Agilent 1100/1200 HPLC system (Agilent Technologies GmbH, Böblingen, Germany), consisting of quaternary pump, autosampler, column oven, and UV detector. From the UV outlet, the flow was directly guided to the NMR system. NMR detection was performed with a Bruker Avance DRX 500 MHz spectrometer (Bruker BioSpin GmbH, Rheinstetten, Germany) equipped with a triple resonance flow probe TXI H(C,N) with z-gradients and an active detection volume of 60 μL. For LCCC-NMR, four different reversed phase columns produced by Macherey Nagel (Düren, Germany) were used (pore sizes [Å]−particle sizes [μm]: 50−5, 100−5, 300−5, 1000−7; each with 25 cm length and 4 mm inner diameter). The mobile phase was 1,4-dioxane (HPLC grade) at a flow rate of 0.3 mL·min−1. Sample concentrations were 14 mg·mL−1 with an injection volume of 0.1 mL. Critical conditions were achieved by varying the column temperature according to refs 12 and 22. The implementation of ERETIC was done by modifying the WET45 pulse sequence and introducing the signal via the15N coil of the NMR flow probe. The artificial signal in the onflow spectra was created at δ = −1 ppm. SEC-NMR was done using three PSS SDV columns (PSS GmbH, Mainz, Germany) with pore sizes of 102 Å, 103 Å, and 105 Å (each had a particle size of 5 μm, a length of 300 mm, and an inner diameter of 8 mm) preceded by a precolumn. THF was used as the mobile phase at a flow rate of 0.8 mL· min−1. The sample concentrations were 7.3 mg·mL−1. A total of 0.1 mL was injected. For relative molar mass determination, the SEC-NMR system was calibrated with PI and PMMA homopolymers (see Table S1 and Figure S1 in the SI).

■

EXPERIMENTAL SECTION Samples. The PI-b-PMMA block copolymers with DPE as the linker (see Figure 1) were synthesized by anionic living polymerization. The synthesis is described in the Supporting Information. Four copolymers were produced with the same 491

dx.doi.org/10.1021/ac402396r | Anal. Chem. 2014, 86, 490−497

Analytical Chemistry

Article

Table 1. Average Compositions, Block Lengths, and Mn of the Copolymers Determined by 1H NMR sample

copolymer composition DPE/PI/PMMA [mol %]

number of monomer units PI/PMMA [−]

average molar mass PI/PMMA Mn [kg·mol−1]

average molar mass copolymer Mn [kg·mol−1]

1 2 3 4

2.8/81.9/15.3 2.0/58.4/39.6 1.6/46.2/52.2 1.2/35.0/63.8

30/5 30/19 29/31 30/53

2.01/0.51 2.03/1.87 1.97/3.15 2.01/5.31

2.70 4.07 5.30 7.50

Figure 2. (a) Elution of PI with varying column temperatures in 1,4-dioxane. (b) Elution of PI and PMMA homopolymers in dioxane under critical conditions of PI as well as a calculated PI calibration based on the chemical compositions of the copolymer (50 °C), ● = PI, ○ = calculated PI, ▲ = PMMA (fitted to third order polynomials (R2 = 0.999)).

using the same intramolecular referencing method. The concept of invisibility of the critical block for LCCC and the influence on the block under SEC conditions will be discussed for the relative and direct methods. The last section will focus on LCCC-NMR using an external reference signal (ERETIC) to show the applicability to polymers without an internal reference for the absolute quantitation of polymers which agrees with weighing. LCCC-NMR vs SEC-NMR Using Common Calibration Standards for Molar Mass Determinations. LCCC conditions were applied only for the PI block by varying the column temperature according to refs 3 and 22. Figure 2a shows the elution behavior of PI homopolymers of different molar masses. The column temperature of 50 °C corresponds to critical conditions indicated by the molar mass independence of PI. The increase of the temperature yields SEC behavior (55 °C line in Figure 2a), whereas the decrease below 50 °C leads to LAC conditions (45 °C line in Figure 2a). The critical conditions of PI were used for the molar mass determination of the PMMA block of PI-b-PMMA. In this case, PMMA homopolymers (as well as the PMMA block) elute in SEC mode and allow for a molar mass calibration as shown in Figure 2b. The four PI-b-PMMA copolymers were measured with LCCC-NMR and SEC-NMR. In addition, a blend of PI endcapped with DPE and PI homopolymer was studied with LCCC-NMR. The onflow LCCC- and SEC-NMR stacked plots of sample 2 are shown in Figure S3, and a contour plot of the blend together with the copolymer sample 4 is shown in Figure S4 of the SI. Figure S4b (SI) shows the main copolymer fraction indicated by the PMMA and PI elution between 23 and 28 min. In addition, a second fraction at 30.5 min is visible in Figure S4b. This fraction could be identified as the precursor of the first block due to unwanted termination reactions. However, the precursor is not the pure PI homopolymer as expected. It is rather PI end-capped with DPE. This structure was verified by

The molar masses obtained by SEC-NMR and LCCC-NMR were characterized with a standard deviation analysis to determine an average error of the molar mass data (see SI). Due to the short repetition times of the onflow NMR detection (for experimental details of the NMR parameters, see the SI) as well as the usage of small copolymers, the resulting errors in signal intensities of all polymer signals were corrected according to the method suggested by Hiller et al.2 The resulting coefficients for both LCCC-NMR and SEC-NMR conditions are shown in Figure S2 in dependence on the PMMA block weights. The correction coefficients of PMMA decrease with increasing PMMA molar mass until MPMMA = 3 kg·mol−1 due to decreasing T1 relaxation times. The relaxation times of PMMA are constant above 3 kg·mol−1. For the given samples, the PI correction coefficients are constant because all PI blocks have the same molar mass. The DPE coefficients show a slight dependence of the PMMA block size due to the direct covalent neighborhood to PMMA. The observed trends were included in the relaxation corrections of all samples.

■

RESULTS AND DISCUSSION In the following three chapters, we will describe the analysis of the four diblock copolymers having identical first PI blocks linked via one DPE unit to different sizes of PMMA blocks. After characterizing the structure of PI-b-PMMA, we will first compare the common LCCC method with the traditional, relative SEC method for determining Mn, Mw, and polydispersities (PDi) using homopolymer standards for the calibration of molar masses vs elution times. Especially, the limits of this relative LCCC method will be demonstrated. Then, we will introduce an intramolecular referencing method using the DPE linker as an intramolecular reference to determine the molar masses of the copolymers. This method will be applied to LCCC-NMR as well as SEC-NMR. In particular, it will be shown that the direct quantification method of LCCC-NMR allows now for the Mn and Mw determination of both blocks. These results will be compared with SEC-NMR 492

dx.doi.org/10.1021/ac402396r | Anal. Chem. 2014, 86, 490−497

Analytical Chemistry

Article

Figure 3. LCCC-NMR and SEC-NMR chromatograms and co-monomer compositions for samples 1 (a,b) and 4 (c,d). Dotted line = DPE chromatogram, dashed line = PI chromatogram, dashed-dotted line = PMMA chromatogram, solid line = copolymer chromatogram, ● = mol % isoprene, ▲ = mol % MMA, green/red linear lines = extrapolated monomer compositions.

Table 2. Average Molar Masses [kg·mol−1] and Polydispersities (PDi) [−] of the PI and PMMA Blocks Using the Homopolymer Calibrations of Figure 2b (LCCC) and Figure S1 (SEC) and Molar Masses of the Copolymers Using the Chemical Compositions of Figures 3 and S5 Together with the Homopolymer Calibrations LCCC-NMRb

SEC-NMRc

PI

PMMA

copolymer

PI

PMMA

copolymer

samplea

Mw/Mn PDi

Mw/Mn PDi

Mw/Mn PDi

Mw/Mn PDi

Mw/Mn PDi

Mw/Mn PDi

1

1.24/1.10 1.13 1.29/1.07 1.21 1.35/1.09 1.23 1.53/1.23 1.24

0.84/0.44 1.90 3.12/1.22 2.55 4.96/1.90 2.6 7.36/3.75 1.96

2.07/1.73 1.19 3.87/2.48 1.56 5.71/3.16 1.80 8.36/5.12 1.63

2.54/2.34 1.09 2.28/2.03 1.12 2.32/2.00 1.16 1.97/1.68 1.17

0.98/0.89 1.10 3.06/2.67 1.15 4.34/3.66 1.19 8.03/6.72 1.19

3.50/3.21 1.09 5.34/4.63 1.15 6.69/5.54 1.2 9.99/8.27 1.21

2 3 4

a Results of sample 3 after mathematical separation of homopolymer. bThe Mw and Mn data of PI are determined via the calculated PI homopolymer calibration of Figure 2b, the data of PMMA via the measured PMMA homopolymer calibration of Figure 2b. cThe Mw and Mn data of PI and PMMA are determined via their homopolymer calibration of Figure S1 and the copolymer data via a copolymer calibration.

comparing the LCCC-NMR elution times of PI end-capped with DPE and the second fraction of the copolymer (Figure S4). The onflow experiments also allow for the calculation of the individual NMR chromatograms (by normalizing the NMR intensities of the aromatic, olefinic and methyl signals to one proton including the relaxation corrections) as well as the chemical compositions in dependence on the elution time (by directly calculating the mol % of each co-monomer from the normalized NMR intensities). Figure 3 shows the LCCC- and SEC-NMR chromatograms and the PI and PMMA chemical compositions of samples 1 and 4. Slightly stronger dependencies of the chemical compositions are found for LCCC allowing also for the separation of the precursor (see second

fraction and strong increase of PI content in Figure 3c). In the case of sample 3, also a high molar mass PMMA component is visible (Figure S5c). The strong changes of the chemical compositions (up to 100%) indicate very heterogeneous samples. In the case of LCCC, the PMMA chromatograms (Figure 3a,c and Figure S5a,c) as well as the PMMA calibration curve of Figure 2b were used for the determination of the average molar masses Mn and Mw of the PMMA blocks with equations S1a and S1b (SI; see Table 2). Molar masses of the PI blocks are not determinable with the relative method due to their elution in critical mode. However, the chemical compositions can also be used to create a PI calibration curve via the PMMA calibration and finally calculating the copolymer data. This is an 493

dx.doi.org/10.1021/ac402396r | Anal. Chem. 2014, 86, 490−497

Analytical Chemistry

Article

Figure 4. LCCC-NMR and SEC-NMR chromatograms and block lengths of the samples 1 (a,b) and 4 (c,d). Dotted line = DPE chromatogram, dashed line = PI chromatogram, dashed-dotted line = PMMA chromatogram, ● = number of isoprene units, ▲ = number of MMA units.

Table 3. Average Molar Masses [kg·mol−1] and Polydispersities (PDi) [−] of the PI and PMMA Blocks As Well As the Entire Copolymers Determined via the Direct Intramolecular Referencing Method Using Figures 4 and S6 and Equations S2−S4 LCCC-NMR PMMA

copolymer

PI

PMMA

copolymer

samplea

Mw/Mn PDi

Mw/Mn PDi

Mw/Mn PDi

Mw/Mn PDi

Mw/Mn PDi

Mw/Mn PDi

1

1.91/1.89 1.01 1.97/1.95 1.01 1.98/1.94 1.02 1.96/1.89 1.04

0.90/0.70 1.29 2.67/2.08 1.28 4.22/3.05 1.38 6.68/5.44 1.23

2.83/2.77 1.02 4.50/4.21 1.07 5.86/5.18 1.13 8.44/7.46 1.13

2.11/2.05 1.03 2.03/2.00 1.02 2.35/2.25 1.05 1.93/1.92 1.01

0.66/0.62 1.07 2.34/2.01 1.16 4.29/3.06 1.40 6.19/5.91 1.05

2.94/2.85 1.03 4.46/4.19 1.06 6.43/5.49 1.17 8.24/8.01 1.03

2 3 4 a

SEC-NMR

PI

Results of the copolymer fraction after simulated separation of sample 3.

lower than those obtained from SEC-NMR. These results are related to the properties of critical conditions where the visible PMMA block affects the copolymer by its SEC behavior and the invisible PI block tends to critical conditions indicated by the increasing PI content with increasing retention times (see Figure 3). Therefore, both mechanisms affect the retention of the block copolymers, and calibrations based on homopolymers can cause incorrectly determined molar masses and polydispersities. It seems that the concept of invisibility of the critical block is not holding up. Recent papers published by Chang et al.46 and Wang et al.47 confirm that the critical block is not completely invisible by studying different block lengths of the invisible block. The concept of invisibility will be discussed in detail in the next chapter. LCCC-NMR vs SEC-NMR Using Intramolecular Referencing for Direct Quantification of Molar Masses. To obtain correct molar masses, a direct method is applied for the

easy approach of determining molar masses of the critical block (see Table 2). Reasonable calibration data can only be calculated for molar masses greater than one monomer unit. In the case of SEC-NMR, the molar masses of both blocks were determined with the individual homopolymer calibration curves of Figure S1 as shown in Table 2. Furthermore, the PI and PMMA chromatograms together with the chemical compositions (Figure 3b,d and Figure S5b,d) and the homopolymer calibrations shown in Figures S1 (SI) were used for molar mass determinations of the entire copolymers with SEC-NMR. The molar masses of the copolymers were analogously determined by the multiple detection procedure by Runyon et al.1 using equations S5 and S6 (SI). The SEC-NMR and the LCCC-NMR results are also listed in Table 2. High polydispersities were obtained from the LCCC-NMR measurements. In particular, the Mn values are significantly 494

dx.doi.org/10.1021/ac402396r | Anal. Chem. 2014, 86, 490−497

Analytical Chemistry

Article

a new microscopic aspect for the concept of invisibility of the critical block and would support the invisibility of the critical block in respect to its chain length independence and chain length distribution within the individual copolymer elution. LCCC-NMR Using External Referencing for Quantification of Molar Masses and Concentration. In this section, the application of the ERETIC technique in LCCCNMR will be discussed. The ERETIC signal (created at δ = −1 ppm) was calibrated with a weighed PMMA homopolymer (Mp = 23.5 kg·mol−1) which was injected into the LCCC-NMR system (cPMMA = 7.4 mg·mL−1, Vinjection = 0.1 mL). The calibration was performed in onflow mode (for more details, see the SI). After calibration, the onflow LCCC-NMR experiments with the ERETIC signal at −1 ppm were performed for the copolymers. The NMR chromatograms of PI, PMMA, and DPE (IDPE/I/MMA,t) could now be converted to molar amounts by using eq 1. IDPE/I/MMA, t nDPE/I/MMA, t = nE · IE (1)

quantification. Because of the linking DPE, an intramolecular referencing can be implemented in the LCCC-NMR and SECNMR measurements. Thus, direct block length and molar mass calculations of all blocks of the copolymers with respect to elution times can be performed for LCCC and SEC by using equations S3 and S4 (SI). The distributions of the average block lengths determined with intramolecular referencing via DPE are shown for LCCC-NMR and SEC-NMR in Figure 4. The results of Mn and Mw are listed in Table 3. According to Figure 4, the SEC-NMR measurements show SEC behavior for all PI as well as PMMA blocks indicated by the decrease of monomer units with increasing retention times. For the LCCC-NMR measurements, only the PMMA blocks show SEC behavior. The number of isoprene units, however, is constant during the copolymer elution for all copolymers (Figure 4a,c and Figure S6a,c in SI). From the block lengths, the average molar masses were determined directly for all blocks and the copolymers (see Table 3) by using equations S2−S4 (SI). All molar masses determined by the direct LCCCNMR method show small polydispersities. Strong deviations of Mn and large polydispersities caused by the relative LCCC method are not observed anymore. Therefore, the direct method based on the internal referencing allows for correct Mn and Mw determinations of the individual blocks as well as the copolymer. Very important conclusions can be derived from these results: (1) Molar masses obtained with LCCC-NMR and SECNMR using the intramolecular referencing method provide reliable and comparable results. (2) The direct LCCC-NMR method should only be preferred if the copolymer needs to be separated from the precursor. (3) When the relative and direct SEC-NMR results were compared, it was found that molar masses determined by the relative procedure are higher than those of intramolecular referencing. The conventional relative method provides inaccurate molar masses due to the usage of homopolymer calibrations and the influence of different hydrodynamic volumes of homo- and copolymers. The intramolecular referencing procedure, however, is independent of these differences. This means for the given block copolymers that the hydrodynamic volumes of the copolymers are higher than the sum of homopolymers with the same chain lengths as the blocks. Consequently, overestimated molar masses are obtained for the relative method. (4) Comparable molar masses for the relative and the intramolecular referencing methods are obtained if samples of the same hydrodynamic behavior as the homopolymers are studied. This is given for sample 5 (polyisoprene end-capped with DPE). The results of Mn and Mw are shown in Figure S7 (SI). (5) The direct SEC-NMR provides SEC behavior for both blocks by showing a decrease of the PI as well as the PMMA chain lengths with increasing elution times, whereas the LCCC-NMR shows a constant PI block length and only SEC behavior for the PMMA block. The constant PI block lengths during elution, however, seem in contradiction to the changing chemical compositions of PI and the comments made above to the validity of critical conditions of PI. However, the change of the chemical compositions is now simply given by the changing PMMA chains lengths at constant PI lengths. The direct LCCC results indicate even a very homogeneous PI block for all samples. This behavior proposes that critical conditions force an averaging out of the PI chain lengths distribution for each PMMA length during the entire copolymer elution and provide a constant average PI chain length which corresponds to the precursor. This is rather

where IE is the ERETIC signal and nE is the ERETIC signal calibrated regarding the molar amount. The results are displayed in Figure S8 in the SI. It should be noted that in the case of PI and PMMA blocks, the results do not correspond to the number of copolymer chains but to the number of monomer units therein. For DPE, however, a direct relation to the number of chains is given due to the fact that every copolymer contains only one DPE unit (nDPE = ncopolymer). The monomer molar masses also allowed for calculation of mass related chromatograms. All resulting curves are shown in Figure S9. By calculating the sums of all molar amounts and masses over the entire elution, the injected amounts and masses of the samples could be determined. The results are shown in Table 4 Table 4. ERETIC Results of Injected Amounts of the Copolymers in Comparison to Weighed Values As Well As Molar Masses Determined via Equation 3 LCCC-NMR results

weighed values

molar masses

sample

ncopolymerchains [μmol]

mcopolymer [mg]

mcopolymer [mg]

Mw/Mn [kg/mol]

1 2 3 4

0.52 0.33 0.23 0.16

1.43 1.38 1.31 1.28

1.55 1.54 1.51 1.44

2.86/2.68 4.42/4.13 6.79/5.41 8.57/7.82

in comparison to the weighed values. It was found that injected amounts calculated by ERETIC referencing nearly match the weighed values. The slightly lower values from the calculations might result from inaccuracies in weighing and dissolving, dead volumes in the chromatographic system, and residual impurities in the solvent, whose NMR signals overlap with copolymer signals and therefore have to be subtracted in the processing procedure. It is evident from the results that substance masses of 1 μg to 10 μg per onflow spectrum as shown in Figure S9 (corresponding to amounts of 10 nmol to 100 nmol in Figure S8, SI) were sufficient for a detailed LCCC-NMR analysis. Each onflow spectrum was accumulated over an elution volume of Ve = 64.1 μL. Because this value almost perfectly corresponds to 495

dx.doi.org/10.1021/ac402396r | Anal. Chem. 2014, 86, 490−497

Analytical Chemistry

Article

the volume of the detection cell in the flow probe (Vprobe = 60 μL), the discussed magnitudes can also be related to the detection volume. Finally, it was also possible to determine the average molar masses Mn and Mw from the ERETIC referencing by using eqs 2, S4a, and S4b (SI). mDPE, t + mI, t + mMMA, t Mcopolymer, t = nDPE, t (2)

(2) Hiller, W.; Hehn, M.; Hofe, T.; Oleschko, K. Anal. Chem. 2010, 82, 8244−8250. (3) Hehn, M.; Hiller, W.; Wagner, T.; Thiel, J.; Pasch, H. Macromol. Chem. Phys. 2012, 213, 401−410. (4) Hiller, W.; Hehn, M.; Hofe, T.; Oleschko, K.; Montag, P. J. Chromatogr. A 2012, 1240, 77−82. (5) Benoit, H.; Froelich, D. Light Scattering from Polymer Solutions; Academic Press: London, 1972. (6) Zimina, T. M.; Kever, J. J.; Melenevskaya, E. Yu.; Fell, A. F. J. Chromatogr. 1992, 593, 233−241. (7) Pasch, H.; Trathnigg, B. HPLC of Polymers; Springer: Berlin, 1998. (8) Pasch, H.; Augenstein, M. Makromol. Chem. 1993, 194, 2533− 2541. (9) Pasch, H.; Augenstein, M.; Trathnigg, B. Macromol. Chem. Phys. 1994, 195, 743−750. (10) Falkenhagen, J.; Much, H.; Stauf, W.; Müller, A. H. E. Macromolecules 2000, 33, 3687−3693. (11) Lee, W.; Cho, D.; Chang, T.; Hanley, K. J.; Lodge, T. P. Macromolecules 2001, 34, 2353−2358. (12) Lee, W.; Park, S.; Chang, T. Anal. Chem. 2001, 73, 3884−3889. (13) Park, I.; Park, S.; Cho, D.; Chang, T.; Kim, E.; Lee, K.; Kim, Y. J. Macromolecules 2003, 36, 8539−8543. (14) Jiang, W.; Khan, S.; Wang, Y. Macromolecules 2005, 38, 7514− 7520. (15) Hatada, K.; Ute, K.; Okamoto, Y.; Imanari, M.; Fujii, N. Polym. Bull. 1988, 20, 317−321. (16) Hatada, K.; Ute, K.; Kashiyama, M.; Imanari, M. Polym. J. 1990, 22, 218−222. (17) Ute, K.; Niimi, R.; Hongo, S.-y.; Hatada, K. Polym. J. 1998, 30, 439−443. (18) Hiller, W.; Hehn, M.; Sinha, P.; Raust, J.-A.; Pasch, H. Macromolecules 2012, 45, 7740−7748. (19) Hiller, W.; Sinha, P.; Hehn, M.; Pasch, H. Prog. Polym. Sci. 2013, [online] http://dx.doi.org/10.1016/j.progpolymsci.2013.10.001. (20) Hiller, W.; Sinha, P.; Pasch, H. Macromol. Chem. Phys. 2007, 208, 1965−1978. (21) Hiller, W.; Sinha, P.; Pasch, H. Macromol. Chem. Phys. 2009, 210, 605−613. (22) Hiller, W.; Pasch, H.; Sinha, P.; Wagner, T.; Thiel, J.; Wagner, M.; Müllen, K. Macromolecules 2010, 43, 4853−4863. (23) Hiller, W.; Sinha, P.; Hehn, M.; Pasch, H.; Hofe, T. Macromolecules 2011, 44, 1311−1318. (24) Sinha, P.; Hiller, W.; Bellas, V.; Pasch, H. J. Sep. Sci. 2012, 35, 1731−1740. (25) Ziarelli, F.; Viel, S.; Caldarelli, S.; Sobieski, D. N.; Augustine, M. P. J. Magn. Reson. 2008, 194, 307−312. (26) Barantin, L.; le Pape, A.; Akoka, S. Magn. Reson. Med. 1997, 38, 179−182. (27) Akoka, S.; Barantin, L.; Trierweiler, M. Anal. Chem. 1999, 71, 2554−2557. (28) Silvestre, V.; Goupry, S.; Trierweiler, M.; Robins, R.; Akoka, S. Anal. Chem. 2001, 73, 1862−1868. (29) Moing, A.; Maucourt, M.; Renaud, C.; Gaudillère, M.; Brouquisse, R.; Lebouteiller, B.; Gousset-Dupont, A.; Vidal, J.; Granot, D.; Denoyes-Rothan, B.; Lerceteau-Köhler, E.; Rolin, D. Funct. Plant Biol. 2004, 31, 889−902. (30) Martínez-Bisbal, M. C.; Monleon, D.; Assemat, O.; Piotto, M.; Piquer, J.; Llácer, J. L.; Celda, B. NMR Biomed. 2009, 22, 199−206. (31) Øystein, R.; Melø, T. M.; Sonnewald, U. NMR Biomed. 2009, 22, 266−271. (32) Heinzer-Schweizer, S.; de Zanche, N.; Pavan, M.; Mens, G.; Sturzenegger, U.; Henning, A.; Boesiger, P. NMR Biomed. 2010, 23, 406−413. (33) Billault, I.; Akoka, S. Instrum. Sci. Technol. 2000, 28, 233−240. (34) Maiwald, M.; Grützner, T.; Ströfer, E.; Hasse, H. Anal. Bioanal. Chem. 2006, 385, 910−917. (35) Billault, I.; Robins, R.; Akoka, S. Anal. Chem. 2002, 74, 5902− 5906.

The results displayed in Table 4 show very good agreement with the direct SEC-NMR and LCCC-NMR results from intramolecular referencing (compare Table 3).

■

CONCLUSION Different referencing methods using intramolecular and extramolecular reference signals were studied for the characterization of block copolymers by online LCCC-NMR and SECNMR. The direct method of intramolecular referencing could be efficiently used to determine accurate molar masses of comonomers with both LCCC-NMR and SEC-NMR. It could be shown that the relative methods resulted in inaccurate molar masses. Remarkably, intramolecular referencing resulted in the same copolymer molar masses regardless of the chromatographic method (LCCC of SEC). The direct method is insensitive against peak broadening or stretching effects. It has to be noted, however, that the direct procedure can only be used when every polymer chain has the same reference group which is detectable during the whole elution period. The detection of the reference group is easily possible with LCNMR as long as the polymers are small (probably up to 20 kg· mol−1). This method gives detailed molecular information on the behavior of the invisible and visible blocks in LCCC. Further, it was shown that the chromatograms obtained in LCCC-NMR could be easily converted into molar amount and mass related curves by the use of the ERETIC method. It was possible to determine the molar amount of copolymer chains at a given retention volume. The limiting magnitude of concentration for online NMR detection could be determined to be 10 nmol to 100 nmol equivalent to 1 μg to 10 μg per onflow spectrum.

■

ASSOCIATED CONTENT

S Supporting Information *

Synthesis of PI-PMMA block copolymers, experimental section, standard deviation analysis for SEC-NMR and LCCC-NMR, eretic experiments, figures, equations, and tables. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Fax: +492317553771. Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS We thank the Deutsche Forschungsgemeinschaft for supporting the project. REFERENCES

(1) Runyon, J. R.; Barnes, D. E.; Rudd, J. F.; Tung, L. H. J. Appl. Polym. Sci. 1969, 13, 2359−2369. 496

dx.doi.org/10.1021/ac402396r | Anal. Chem. 2014, 86, 490−497

Analytical Chemistry

Article

(36) le Grand, F.; George, G.; Akoka, S. J. Magn. Reson. 2005, 174, 171−176. (37) le Grand, F.; George, G.; Akoka, S. J. Agric. Food Chem. 2005, 53, 5125−5129. (38) Remaud, G. S.; Silvestre, V.; Akoka, S. Accredit. Qual. Assur. 2005, 10, 415−420. (39) Michel, N.; Akoka, S. J. Magn. Reson. 2004, 168, 118−123. (40) van Lokeren, L.; Kerssebaum, R.; Willem, R.; Denkova, P. Magn. Reson. Chem. 2008, 46, S63−S71. (41) van Lokeren, L.; Kerssebaum, R.; Willem, R.; Denkova, P. Magn. Reson. Chem. 2011, 49, 137−139. (42) Ziarelli, F.; Caldarelli, S. Solid State Nucl. Magn. Reson. 2006, 29, 214−218. (43) Ziarelli, F.; Viel, S.; Sanchez, S.; Cross, D.; Caldarelli, S. J. Magn. Reson. 2007, 188, 260−266. (44) Franconi, F.; Chapon, C.; Lemaire, L.; Lehmann, V.; Barantin, L.; Akoka, S. Magn. Reson. Imaging 2002, 20, 587−592. (45) Smallcombe, S. H.; Patt, S. L.; Keifer, P. A. J. Magn. Reson. A 1995, 117, 295−303. (46) Lee, S.; Lee, H.; Thieu, L.; Jeong, Y.; Chang, T. Macromolecules 2013, 46, 9114−9121. (47) Yang, X.; Zhu, Y.; Wang, Y. Polymer 2013, 54, 3730−3736.

497

dx.doi.org/10.1021/ac402396r | Anal. Chem. 2014, 86, 490−497