WO2001025784A1

WO2001025784A1 - Method for testing the degradation of polymeric materials

Info

Publication number: WO2001025784A1
Application number: PCT/US2000/027735
Authority: WO
Inventors: Joseph A. Gardella, Jr.; Jiaxing Chen; Norma Hernandez De Gatica; Joo-Woon Lee
Original assignee: Research Foundation Of State University Of New York
Priority date: 1999-10-06
Filing date: 2000-10-06
Publication date: 2001-04-12
Also published as: AU7999400A

Abstract

The present invention provides a novel method for monitoring the degradation of biodegradable polymers including, but not limited to, polyesters and polyanhydrides, at the molecular level, to provide detailed information on degradation rates, exploring degradation kinetics and mechanism, by using Time-of-Flight Secondary Ion Mass Spectrometry (ToF SIMS) measurements. The method comprises (1) preparing polymer specimen to predetermined dimensions; (2) initiating the desired degradation reaction in vitro; (3) monitoring the degradation process with a ToF SIMS instrument by measuring the degradation products generated; (4) conducting data reduction which can provide detailed information on degradation rates, kinetics and mechanisms of the degradation.

Description

METHOD FOR TESTING THE DEGRADATION OF POLYMERIC MATERIALS

Research Funding Source

This invention was made with government support under grant CHE9704996 from the National Science Foundation. The government has certain rights in the invention.

Claim of Priority

This application claims priority to U.S. provisional patent application no. 60/157,964, filed on October 6, 1999.

Field of the Invention

The present invention relates generally to the field of biodegradable polymers. More particularly, the present invention provides a method for determining the reaction kinetics of biodegradable polymers.

Description of Related Art Synthetic biodegradable polymers have been used in clinical applications for decades. Some relevant applications include surgical implants, wound healing materials, absorbable sutures and drug delivery devices. Among issues important in developing biomedical applications based on polymer biodegradability are the properties of degradation (such as rate, mechanism, by products, etc.) of the polymer material. The study of hydrolytic degradation of biodegradable polymers has been a research focus in the past few decades with in vivo investigations of biopolymer implants being the major clinical investigation method. Direct monitoring of the weight loss of polymer implants and histological the hydrolytic degradation. In Addition, a drawback of this method is that it is very time consuming.

For the in vi tro investigation of hydrolytic degradation of biodegradable polymers, many bulk characterization have been developed. Properties such as tensile strength, thermal properties, mass loss, and decrease m molecular weight have been measured. Techniques used include differential scanning calorimetry (DSC) , gravi etry, gel permeation chromatography (GPC) , size exclusive chromatography (SEC) , FTIR, NMR, X-ray diffraction, and laser diffractometry.

Among the surface sensitive microscopic and spectroscopic techniques, such as scanning electron spectroscopy (SEM) and atomic/scanning force microscopy have become important means in studying biodegradation of polymers. The surface microscopic techniques, however, do not provide chemical compositional or structural information. Thus, conventional techniques do not provide information at the molecular level, which is essential for custom design of polymer structure to obtain the desired degradation rate, in addition to fundamental understanding of the reaction mechanism. Thus, there is a pressing need to develop powerful and fast methods for evaluating the degradation kinetics of biodegradable polymers .

Summary of the Invention The present invention provides a novel method for determining reaction kinetics of biodegradable polymers by using Ti e-of-Flight Secondary Ion Mass Spectrometry (ToF SIMS) measurements on polymeric materials. The method comprises (1) preparing polymer specimen to predetermined dimensions; (2) initiating the desired degradation reaction in vitro; (3) monitoring the degradation process with a ToF SIMS instrument by measuring the degradation products generated; and (4) conducting data reduction which can provide detailed information on degradation rates, kinetics, and mechanisms of the degradation.

Brief Description of the Drawings Figure 1 is a representation of a high mass portion of the ToF SIMS spectra of hydrolyzed PGA.

Figure 2 is a representation of a high mass portion of the ToF SIMS spectra of hydrolyzed PLA.

Figure 3 is a representation of a high mass portion of the ToF SIMS spectra of hydrolyzed PLGA 50:50 copolymer.

Figure 4 is a representation of the ToF SIMS spectra of hydrolyzed PLGA from 800 D to 1000D. Number marked peaks are molecular ion peaks. Figure 5 is a comparison of a group of molecular peaks in the ToF SIMS spectra with the theoretically calculated mass spectra.

Figure 6 is a representation of a high mass portion of the ToF SIMS spectra of hydrolyzed PSA. Figure 7 is a representation of a high mass portion of the ToF SIMS spectra of PFS 20:80 copolymer before hydrolysis treatment.

Figures 8 is a representation of a high mass portion of the ToF SIMS spectra of hydrolyzed PFS 20:80 copolymer.

Figures 9 is a representation of a high mass portion of the ToF SIMS spectra of hydrolyzed PFS 50:50 copolymer. Figure 10 is a comparison of the low mass portion (90 D to 103D) of the ToF SIMS spectra of PFS copolymer before and after hydrolysis.

Figure 11 is a plot of log molecular ion peak intensity of hydrolyzed polyanhydrides versus degree of polymerization of the hydrolysis products.

Detailed Description of the Invention

The present invention discloses a novel method for determining reaction kinetics of biodegradable polymers. The method can be used for in vi tro studies of polymeric biomaterials and significantly reduce the needs of in vivo studies. The method comprises detecting and identifying biodegradation products of the polymers in ToF SIMS spectra as intact molecular ions, and the molecular weight distribution of the degradation products .

Data is presented for the hydrolytic degradation of six biodegradable polymers, involving two important classes of biodegradable polymers, in particular polyesters and polyanhydrides, and both homopolymers and random copolymers, using ToF SIMS. Upon hydrolytic degradation of the polymers, low molecular weight oligomers generated during the hydrolytic degradation can be directly detected in ToF SIMS spectra. The following examples are presented for illustrative purposes and are not to be construed as being restrictive.

Experimental

Poly (glycolic acid) (PGA) was obtained from the Davis & Geek Division of American Cyanamid Company; Poly (1-lactic acid) (PLA) (weight average molecular weight 93,000) and poly (dl-lactic-co-glycolic acid) (PLGA) (50:50, molecular weight 50,000 - 75,000) were purchased from Sigma Chemical Company (St. Louis, MO); Poly (fumaric-co-sebacic acid) (PFS) (50:50, average molecular weight 3,046, 20:80, average molecular weight 6,500) and poly(sebacic acid) (PSA, average molecular weight 12,000) were supplied by Brown University; and Physiological solution ISOTON^1* II was purchased from Coulter Diagnostics in Hialeah, FL.

PGA, PLA and PLGA samples were prepared by elt- pressing on aluminum foil. Prior to each melt-press, thick PGA plates were cut to small pieces and washed with an ultrasonic cleaner (from Branson Cleaning Equipment Company, model B-52) in hexane and chloroform for 10 minutes each. Aluminum foil was pre-cleaned with chloroform. PLA and PLGA were used as received.

Samples were pressed at about 200°C to about 1 mm of thickness. The aluminum foil was peeled off and hydrolysis treatment was conducted immediately.

Polyanhydride samples were prepared by melting polymers on aluminum foil at their melting temperatures. Polymer samples, about 1mm thick, were pre-cleaned in hexane and vacuum-dried prior to melt. The aluminum foil was peeled off for PSA samples.

The hydrolytic degradation of all polymers was carried out in a physiological buffer solution, ISOTOKT II (pH =7.4), at 37.0^EC. Each sample was immersed in a separate vial prefilled with 14 ml of ISOTOINP II solution and sealed airtight. The reaction vials were immersed in a temperature bath for pre-determined periods depending on the sample' s sensitivity to hydrolytic degradation and other properties such as molecular weight and hydrophobicity of the polymer. The hydrolysis time for each polymer was chosen so that the molecular ions of hydrolysis products were observed with good peak intensity. The hydrolysis times of all polymer samples used m this study are listed m Table 1.

Table 1. Initial Molecular Weight and

Time of Hydrolysis Reactions

Polymer Molecular Weight Hydrolysis Time (m hours)

PGA NA 1 PLA 93k 30

PLGA 50-75k 24

PSA 12k 24

PFS 20:80 6.5k 3

PFS 50:50 3k 2

Samples after the hydrolysis treatment were vacuum- dried at ambient temperature, and stored m sealed vials filled with argon until ToF SIMS analysis was performed. ToF SIMS analysis was conducted on a Physical Electronics 7200 time-of-flight secondary ion mass spectrometer equipped with a pulsed cesium ion gun and a channel plate detector. The primary ion gun was operated at 8keV m all spectral acquisitions. The static mode was used m all acquisitions with the primary ion current of 0.3 pA. The pulse width of primary ion current was 1.0 ns . The total ion dosage m each spectral acquisition did not exceed 1 x 10¹¹ ion/cm². An electron neutralizer was operated during all spectral acquisitions m pulse mode at low electron energy with a target current under IμA for charge compensation. A time resolution of 1.25 ns per step was used for good signal-to-noise ratio at high m/z range. Data reduction was performed using Physical Electronics TOFPak software (version 2.0).

EXAMPLE 1 ToF SIMS Results of Hydrolyzed Polyesters.

Ions from polymer chain fragments, with exponentially deceasing intensity, are normally observed in the high mass range ToF SIMS spectra of thick film polymer samples, unless the sample film is prepared as monolayers deposited on metal substrates. Oligomeπc ion distributions are not normally observed for thick films due to the entanglement of the long chain molecules m polymer samples. Random chain scission occurs when samples are bombarded by primary ions, which transfers energy to polymer chains near the surface.

This process generates fragment ions m ToF SIMS spectra with the characteristic of exponentially decreased intensity with few meaningful peaks due to the decreased possibility of producing high mass fragments. Upon hydrolysis, polymer chain-lengths are reduced gradually until the oligomers become small enough to desorb from the polymer surface and dissolve m the surrounding liquid phase, where they continue to hydrolyze; yielding monomers as the ultimate reaction products. On the surface of a degraded polymer, the entanglement of the degradation-generated oligomers is greatly reduced because of the decrease in molecular chain-length. During ToF SIMS measurements, low molecular weight oligomers become easier to desorb from the sample surface upon the bombardment of primary ions. Therefore, when degradation occurs at polymer sample surfaces, intact molecular ions of the degradation products can be observed m relatively high mass range of ToF SIMS spectra. ToF SIMS Results of Hydrolyzed PGA.

Figure 1 shows the high mass portion (600 D to 2000 D) of ToF SIMS spectra of PGA hydrolyzed for one hour. Before the hydrolysis treatment, essentially nothing can be observed in this range except a noisy background. Upon hydrolysis, a peak pattern characterized by the differences due to the mass of the repeat unit of PGA was observed.

All the ions of the major peaks in Figure 1 have the structure of [nG+H₂0+Na] ⁺, where G stands for the repeat unit of PGA. This indicates that the ions detected from the hydrolyzed PGA sample are in the intact molecules of the hydrolytic degradation product. The sodium ion comes from the external buffer treatment solution and participates in the process of secondary ion formation as an lonization assisting agent. It will be seen that m all samples studied, sodium plays an important role in the lonization process and all species detected in this series contain at least one sodium ion each. It was also observed that, when the concentration of potassium is high enough to promote secondary ion formation of hydrolysis products, a set of molecular ion peaks associated with potassium could be present simultaneously with the series of peaks associated with the sodium ion. This will be discussed below.

Molecular ions of up to 33 repeat units were observed in ToF SIMS spectra of some samples of hydrolyzed PGA. Small oligomers detected in this study can be traced down to the final hydrolysis product, for example the single glycolic acid molecule associated with one sodium ion. Table 2 lists the full range of molecular ions have been detected m this study, with the ions shown in Figure 1 listed in shaded cells. Table 2 Hydrolysis Products of PGA and

PLA Observed in ToF SIMS Spectra

Number of m/z monomers PGA PLA composition nG + H₂0 nL + H_:0

1 99 113

2 157 185

3 215 257

4 273 329

5 331 401

6 389 473

7 447 545

8 505 617

9 563 689

10 621 761

11 679 833

12 737 905

13 795 977

14 853 1 049

15 911 1121

16 969 1193

17 1027 1265

18 1085 1337

19 1143 1 409

20 1201 1481

21 1259 1553

22 131 7 1 625

23 1375 1 697

24 1433 1 769

25 1491 1841 Number of m/z monomers PGA PLA omposition nG + H₂0 nL + H,0

26 1549

27 1 607

28 1 665

29 1 723

30 1 781

31 1839

32 1 879

33 1955

34 2013

35 2071

36 2129

37 2187

38 2245

39 2303

40 2361

41 2419

42 2477

43 2535

44 2593

45 2651

Peaks in shaded cells are shown in Figure 1 and 2.

In addition to the wide distribution of molecular ion peaks, a crest of the molecular ion peaks can also be seen. This crest (1) was first observed in the one hour hydrolysis sample spectra (Figure 1) at 1400 D to 1500 D, and (2) became more pronounced when it moved to the low mass range gradually as the hydrolysis time increased. This crest represents the most probable molecular weights of hydrolysis products at the particular reaction time. The hydrolytic degradation kinetics can be explored using the data from the ToF SIMS analysis.

ToF SIMS of Hydrolyzed PLA.

PLA has the same main chain backbone as PGA plus a methyl group as a side chain. The presence of the methyl group, however, significantly changes the properties of the ester carbon as well as the bulk polymer properties such as morphology and hydrophobicity . These changes are reflected in the characteristic rates of hydrolytic degradation and consequently in the ToF SIMS spectra of hydrolyzed samples. Figure 2 shows the high mass portion (from 500 D to 2000 D) of ToF SIMS spectra of PLA disc samples hydrolyzed for 30 hours. The star marked peaks are the most intense peak in each repeat pattern. As in the spectra of PGA, each repeat pattern corresponds to one repeat unit of PLA. The intervals between each of the star marked peaks are 72.02 m/z, exactly the mass of one PLA repeat unit.

Molecular ion peaks are observed from the surface of hydrolyzed PLA from the final hydrolysis product (for example, the single lactic acid molecule) up to the oligomer with 25 repeat units of PLA. The intensity of the low mass species (not shown in the figures) become lower and lower quickly, this is likely due to the increased diffusibility of the low mass hydrolysis products from the solid sample surface to the hydrolysis solution and the increased solubility as the molecular weight is reduced. A remarkable difference between PLA and PGA is the hydrolytic degradation rate. The spectra shown in Figure 2 are PLA hydrolyzed for 30 hours under the same hydrolysis conditions as that for PGA. For PLA samples hydrolyzed in shorter times, good peak patterns of hydrolysis products could not be observed. In addition, there is no peak crest observed at this hydrolysis time, as seen in the ToF SIMS spectra of hydrolyzed PGA (Figure 1) . The intensity of molecular ion peaks is exponentially decreasing as the m/z increases.

ToF SIMS of Hydrolyzed PLGA.

Figure 3 shows the high mass portion (400 D to 2000 D) of the ToF SIMS spectrum of PLGA 50:50 random copolymer hydrolyzed for 24 hours. The pattern of the spectra is obviously more complicated than both the PGA and PLA spectra. Each of the peaks in Figure 3 consists of a group of peaks (see the inset in Figure 3) . The intervals between the groups are essentially 14 D indicating the repeat pattern is governed by the structural difference between monomeric lactic and glycolic acids. In addition, the most intense peak in each group shifts gradually towards lower m/z so that the overall interval of the repeat pattern is not the exact m/z of one CH₂ group over the full range shown in Figure 3. This complicated pattern can be understood by considering all possible compositions of hydrolysis productions of this random copolymer. Table 3 tabulates the m/z values of all possible molecular ion compositions in terms of m/z for low molecular weight oligomers of the hydrolysis products up to 15 PLA repeat units and 12 PGA repeat units.

LO LO These ions represent the structure of intact molecules, which would be then cationized with at least one sodium ion each. All molecular ions listed in Table 3 are observed in the ToF SIMS spectra except those with high composition ratios of PGA over PLA repeat units. This is because PGA is more sensitive to hydrolysis than PLA. To illustrate this fact, ions in the framed cells (oligomer molecular ions from 800 D to 1000 D) are shown and marked with m/z values in Figure 4. It shows that most peaks shown in the spectra are molecular ion peaks with the most intense peak shifts to the left in each cluster due to the composition of the molecular ions. In fact, the most intense peak in each cluster is always corresponding to the ion with the smallest number of glycolic acid repeat units and the largest number of lactic acid repeat units. The peak intensities for the ions in the shaded cells of Table 3 consisting of larger ratio of glycolic acid repeat units than the ions to their left, become too low to be detected.

The relative intensity of molecular ion peaks also serves as an indication of the relative hydrolysis rate of the two components of the copolymer. PGA hydrolyzes faster than PLA as observed in the homopolymers of PGA and PLA; therefore, fewer PGA repeat units remain in the hydrolysis products. No oligomers with only PGA repeat units were observed while oligomers with pure PLA were observed as the most intense peak in its group (peaks of 833,905 and 977 D in Figure 4, for example). Figure 5 exemplifies the composition of the molecular ion peaks by comparing one of the peak groups with the theoretically calculated spectra. Figure 5a is a small portion of the spectrum shown in Figures 3; and Figure 5b is the corresponding molecular ion peak region theoretically calculated using Googly (Copyright 1994, Andrew Proctor) , which taking the elemental isotopic abundance into account. The match m peak position and relative intensity between the experimentally recorded spectra and the theoretically predicted one supports the assignment of the peaks. Figure 4 indicates that all major peaks in the ToF SIMS spectra of hydrolyzed PLGA copolymer are intact molecular ions of the hydrolysis products .

PGA is the simplest biodegradable poly (α-hydroxy acid) with high crystallinity and hydrophilicity . PLA s only one methyl group different from the structure of PGA as the side-chain on the "-carbon, which causes a remarkable change in its properties from PGA, in addition to the formation of two monomeric enantiomeric structures, and copolymers of different tact cities.

For example, the crystallinity of both P(d)LA and P(1)LA is lower than PGA and the presence of the methyl group in PLA significantly decreases its reactivity toward ester hydrolysis mechanism due to the electron donating effect, resulting in the global decrease in hydrophilicity. As a result, the increased hydrophobicity can reasonably explain the relatively slower hydrolytic degradation of PLA and makes PLA dissolved well in common organic solvents m contrast to PGA, which is soluble only in hexafluoroisopropanol .

Hence, these properties can influence not only each own hydrolytic degradation properties, but also the fragmentation process upon the bombardment of the primary ions. Therefore, the difference m each hydrolytic degradation rates between PLA and PGA can be supported by the observation from ToF SIMS spectra of hydrolyzed PLGA 50:50 copolymer. As shown m Figure 4, peaks consisting of more PLA repeat units and less PGA repeat units are always even more intense than those consisting of more PGA repeat units; indicating that PLA segments are less reactive to hydrolysis than PGA segments. The overall trend of molecular ion peak intensities of the hydrolyzed PLGA copolymer is similar to that of hydrolyzed PLA spectra (Figure 2).

EXAMPLE 2 ToF SIMS of Hydrolyzed Polyanhydrides.

Polyanhydrides are significantly different from polyesters in that the anhydride linkage in the backbone is more vulnerable to attack by water than the ester bond. This leads to faster hydrolysis rates for polyanhydrides and causes a narrow molecular weight distribution of the hydrolysis products. Figure 6 shows the ToF SIMS spectra of hydrolyzed PSA from 400 D to

1200 D. Molecular ion peaks are the most dominant one in each repeat pattern, indicating that the anhydride bond is far easier to break than the alkyl chain. Intact molecular ions observed are from the single sebacic acid molecule up to the oligomer of six PSA repeat units. The single sebacic acid molecule, which is the first member in the series of hydrolysis products, however, is small enough to be dissolved in the hydrolysis solution. Hence, very few of them stay on the surface of samples after the hydrolysis experiments. Therefore, the first significant molecular ion peak is the one that consists of two PSA repeat units. All observed oligomers of the hydrolysis products are listed in Table 4 in the form of actually observed ions, consisting of the intact molecules attached with a sodium ion. Table 4. Molecular Ion Peaks Observed in

Hydrolysis Products of Polyanhydrides

* Molecular ion peaks of the final products are very low intensity, not shown in the figures.

Remarkably different from the polyesters, the intensity of molecular ion peaks drops quickly and exponentially. It can be seen from the spectra in Figure 6 that there would be no species larger than the six-repeat-unit oligomer detectable on the sample surface .

ToF SIMS of Hydrolyzed PFS Copolymers .

Two random copolymers of PFS have been studied, m particular 50:50 and 20:80 by weight percentage of fumaric acid to sebacic acid ratio. The initial molecular weight of 50:50 PFS sample is about 3000 by number average molecular weight, and that of 20:80 PFS sample is about 6000. Considering the molecular weights of the repeat units being 98 and 184 for furmaric acid and sebacic acid, respectively, the degree of polymerization is quite low for both copolymers, approximately 20 for the 50:50 copolymer and 36 for the 20:80 copolymer. It was expected that unreacted oligomers might be detected in the medium mass range of ToF SIMS spectra of unhydrolyzed samples. Figure 7 shows the ToF SIMS spectra of 20:80 copolymer before hydrolysis from 350 D to 1050 D. As it is expected, significant ion series were detected up to 1000 D. The ion sequence of 467 D, 651 D, 835 D and 1019 D ( framed- number marked in Figure 7) has the composition of [F+nS+H]⁺, in which F and S represent the repeat unit of fumaric acid and sebacic acid, respectively. However, in addition to the fragment ion peaks, the molecular ion peak series of 409 D, 593 D, 777 D and 961 D are also present. The composition of this series conforms to the ion structure [nS+H₂0+Na] ⁺, indicating that the copolymer has already partially hydrolyzed during storage. Note that the relative intensity of the two series changes as the m/z increases. The fact that the relative intensity of the molecular ion peak series to the fragment series increases as the m/z increase indicates that the distribution of molecular oligomer ions is independent from the fragment ion distribution.

Upon hydrolysis treatment, the fumaric acid repeat unit could not be detected from any hydrolysis products. Figure 8 shows the ToF SIMS spectra of 20:80 copolymer from 400 D to 1200 D. The spectra of hydrolyzed PFS copolymer are almost the same as that of hydrolyzed PSA, indicating that fumaric acid component in the random copolymer chain sequence is far more sensitive to a hydrolysis environment, and hydrolyzes faster than the PSA sequences. The marked peaks in Figure 8 have exactly the same ion composition as the hydrolysis products of PSA. The difference between the spectrum and that of the product of hydrolyzed PSA is that the largest molecule detected in 20:80 PFS copolymer has five sebacic acid repeat units while the largest molecule detected in PSA has six sebacic acid repeat units, in spite of the shorter hydrolysis time for the copolymer samples. This is an additional indication of the faster hydrolysis property of the PFS copolymer compared to the homopolymer of PSA.

Similar results were observed for the 50:50 PFS copolymer. Figure 9 shows the ToF SIMS spectra of the hydrolyzed sample from 400 D to 1200 D. Similar to the 20:80 sample, only low molecular weight oligomers of PSA were observed and products are more narrowly distributed to lower molecular weights. The largest oligomer molecule observed in the two hour hydrolyzed sample of this copolymer has only four sebacic acid repeat units. In addition, the intensity of the molecular ion peaks decreases much faster than the 20:80 copolymer and the homopolymer of PSA, indicating higher hydrolysis rate is associated with higher fumaric acid content. Although there was no fumaric component detected in the hydrolyzed sample of both the two PFS copolymers, fragment ions from fumaric acid repeat units were indeed detected in all cases, indicating that the fumaric acid repeat unit is released from the polymer chain sequence as a whole unit during the hydrolytic degradation. Figure 10 shows the fragment peaks of fumaric acid single unit before and after hydrolysis for two hours. Peaks of 97 D, 98 D and 99 D are the ions of [F-H]⁺, F⁺ and [F+H]⁺, respectively. The increased relative intensities of F+ to other peaks may also indicate the contribution of hydrolytic cleavage of fumaric acid bonds .

Both PSA and PFA (poly (fumaric acid)) are highly crystalline materials. It has been determined that the crystallinity of homopolymers of PSA and PFA are 66%, respectively. The crystallinity of their copolymers decreases depending on the composition of the copolymer, but it is no lower than 38% for all compositions. The samples of the polyanhydrides studied in this work, therefore, cannot be made by solution-casting. One of the concerns for the melt-cast sample preparation procedure is the possibility of oxidation or cross- linking of the double bond in fumaric acid at elevated temperatures. There is no evidence found, however, that the double bond in fumaric acid has been severely changed. The intense fragments of fumaric acid at 97 D, 98 D and 99 D, corresponding to [F - H]⁺, F⁺ and [F + H]⁺, are evidence of the existence of an abundance of unreacted fumaric acid structures (Figure 10a). This structure was also detected in high intensity after the hydrolysis treatments (Figure 10b) , suggesting the basic repeat units of fumaric acid was not changed during the hydrolysis reaction either.

However, fragment ions containing multiple fumaric acid repeat units were never detected in this study. In the ToF SIMS spectra of PFS samples without hydrolysis treatments, only one fumaric acid repeat unit was detected in fragment sequences containing fumaric acid, whereas fragments with up to five sebacic acid repeat units were detected. The molecular ion peak series of hydrolysis products were found in both spectra of 20:80 and 50:50 copolymer samples before hydrolysis treatments. These molecular ion peak sequences consist of only sebacic acid monomers, suggesting that the anhydride bond of fumaric acid is more sensitive to hydrolysis than the anhydride bond of sebacic acid. There are two factors each may play an important role m this issue. One is the hydrophilicity. Sebacic acid contains a highly hydrophobic aliphatic structure while the structure of fumaric acid is highly hydrophilic. This may result in the hydrolytic degradation during storage to occur selectively at the fumaric acid sequence. The other factor is the conjugative property of the fumaric acid structure. With two carboxylic acid groups bridged by a double bond, fumaric acid forms a conjugated structure. This conjugated system increases the reactivity of the carboxylic acid carbon towards nucleophilic reactions.

The differences in hydrolytic degradation rates among the three polyanhydrides can be seen by the hydrolysis time and the hydrolysis products illustrated in the ToF SIMS spectra. Table 4 lists all molecular ions detected in the hydrolyzed samples of all three polyanhydrides. The largest molecule of PSA hydrolysis products has six sebacic acid repeat units while the 20:80 copolymer has five and the 50:50 copolymer has four, although shorter hydrolysis time were used for the PFS copolymers. Furthermore, the intensities of the most intense molecular ion peaks are about 3500, 2000 and 800 for 50:50, 20:80 and the homopolymer of PSA, respectively. Therefore, the molecular ion peaks decrease faster for copolymers that have higher fumaric acid content.

EXAMPLE 3

Information for Kinetics Analysis.

As mentioned above in Example 1, a crest of the molecular ion peaks exists which grows and moves towardthe low mass end when the hydrolysis time increases. Under the assumption that this crest represents the most probable molecular weight distribution of the hydrolysis products, the average molecular weight of the hydrolysis products can be calculated from the ToF SIMS spectra. The average molecular weight obtained from ToF SIMS is a function of hydrolysis time. A linear relationship between the apparent molecular weight and the hydrolysis time has been observed. This observation suggested that the ToF SIMS spectra of hydrolyzed samples carry information about the hydrolytic degradation process of the polymer, which can be used in kinetics and mechanism analysis of hydrolytic degradation of the polymer. Based on this assumption, ToF SIMS studies of the degradation kinetics can be carried out.

As discussed above, the size of the largest molecule of the hydrolysis products of polyanhydrides decreases as the content of fumaric acid m the copolymer increases, and the molecular ion peak intensity decrease faster accordingly. Obviously, this is directly related to the hydrolytic degradation property of the polymer. It is not a surprise that the intensity of the molecular ion peaks decreases exponentially. When the logarithm of the intensity of the molecular ion peak is plotted versus the degree of polymerization, a good linear relationship exists, as shown in Fig. 11. It should be noted that, there are only three and four points in the plot (Figure 11) for the 50:50 and 20:80 copolymers, respectively, due to the fast degradation rates of these copolymers. However, it is important to notice that the slope of the plot reflects how fast the molecular ion peak intensity decreases, and where the line crosses with the line of Log(y)=0, which indicates the molecule of the size marked by the cross point could not practically exist anymore. The slopes of the straight lines have a proportional correlation with the hydrolytic degradation rates. Therefore, the hydrolytic degradation rate should be able to be quantitatively described by this parameter. More data must be collected, however, before quantitative relationships for the hydrolytic degradation reactions can be established.

The hydrolytic degradation of six biodegradable polymers, involving two important classes of biodegradable polymers, m particular polyesters and polyanhydrides, and both homopolymers and random copolymers, has been studied using ToF SIMS. It has been demonstrated that upon the hydrolytic degradation of biodegradable polymers, low molecular weight oligomers generated during the hydrolytic degradation can be directly detected by ToF SIMS spectra for all biodegradable polymers. The oligomers desorb from the sample surface upon the bombardment and ionization, in the form of intact molecules, usually attached with an alkali metal ion. In most cases, the molecular ion peak is the most intense peak in each repeat pattern of ToF SIMS spectra of hydrolyzed polymers.

Based on the observations demonstrated above, it can be concluded that the direct observation of hydrolytic degradation products using ToF SIMS is common to all hydrolytic degradation of biodegradable polymers. Moreover, the ToF SIMS spectra have shown that the ToF SIMS data of hydrolyzed polymer provide the information for kinetics analysis of hydrolytic degradation, which is known to those skilled in the art. The study of biodegradation of polymeric materials using ToF SIMS provides a fast and direct access to the degradation products. The time scale of each observation in ToF SIMS study of the degradation of biodegradable polymers is reduced to hours from weeks or months with the conventional techniques.

From the foregoing, it will be obvious to those skilled in the art the various modifications in the above-described methods, and compositions can be made without departing from the spirit and scope of the invention. Accordingly, the invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. Present examples and embodiments, therefore, are to be considered in all respects as illustrative and not restrictive, and all changes which come within the meaning and range of equivalency of the specifications are therefore intended to be embraced therein.

Claims

We claim :

1. A method for determining reaction kinetics of a biodegradable polymer comprising the steps, which are in no particular order, of: selecting a biodegradable polymer; initiating hydrolytic degradation of the biodegradable polymer in vi tro; monitoring the biodegradation process of the biodegradable polymer with a time-of-flight secondary mass spectrometry instrument to measure oligomers generated during the hydrolytic degradation which provides information for kinetics analysis of the biodegrable polymer within a predetermined time frame; calculating the reaction kinetics of the biodegradable polymer.

2. The method of claim 1 wherein the polymer is shaped to a predetermined size.

3. The method of claim 2 wherein the predetermined size is up to 1 millimeter thick.

4. The method of claim 1 wherein the hydrolytic degradation occurs by immersing the polymer in a polymeric degradation solution.

5. The method of claim 4 wherein the polymeric degradation solution is various saline buffers containing phosphate, acetate, carbonate, or biphthalate having a pH range between about 4.0 and about 10.0.

6. The method of claim 1 wherein the time-of-flight secondary mass spectrometry instrument has a pulsed cesium ion gun and a channel plate detector.

7. The method of claim 1 wherein the biodegradable polymer is homo-/random co-polyesters consisting of from - to ε-hydroxy acids as a monomeric repeat unit and homo-/random co-polyanhydrides consisting of sebacic or fumaric acids as a monomeric repeat unit at least.

8. The method of claim 1 wherein the predetermined time frame is within one week from the step of initiating hydrolytic degradation of the biodegradable polymer.

9. The method of claim 1 wherein the predetermined time frame is within one day from the step of initiating hydrolytic degradation of the biodegradable polymer.