WO1991011695A2

WO1991011695A2 - Measurement of deformation

Info

Publication number: WO1991011695A2
Application number: PCT/GB1991/000118
Authority: WO
Inventors: John Lawrence Stanford; Robert Joseph Young; Richard John Day; Xiao Hu
Original assignee: The University Of Manchester Institute Of Science & Technology
Priority date: 1990-01-26
Filing date: 1991-01-28
Publication date: 1991-08-08
Also published as: WO1991011695A3; GB9001845D0

Abstract

A method of determining a deformation condition initially comprises forming a tractable product into a desired configuration. This product being one which is, or is a precursor to, a polymeric material having chains including Raman active bonds which are capable of stretching but not rotation and which provide at least one diagnostic band in the Raman spectrum of a frequency dependent on the deformation condition of the material. The Raman active material is preferably a copolymer having poly(diacetylene) chains in a hard segment phase. Raman excitation radiation is directed to said material and the frequency (or a representation thereof) of said at least one diagnostic band is detected. This value is compared with the value obtained for the corresponding band of the material in a reference condition. The deformation condition is determined from this comparison.

Description

The present invention relates to the measurement of deformation in materials using Raman spectroscopy. The term "deformation" used herein is to be interpreted as covering any form of stress or strain. The term "material" used herein is to be interpreted as covering any homogenous or heterogeneous material in isolation or in combination with other different materials.

It is well-known that strain in materials can be measured using conventional resistance strain gauges. They are normally attached to the surface of a material under investigation and connected by wires to electronic measuring instruments. They are however limited to measuring the average strain in regions of the order of more than 1 mm².

It is known that stress at a particular point on a silicon semi¬ conductor material may be measured by Raman spectroscopy, see for example GB-A-2 206 688. This technique makes use of the fact that the exact position of at least one band (the diagnostic band) in the Raman spectrum is dependent on the stress in the material. The method is effected by directing Raman excitation radiation (usually a laser) to the point of interest on the semi-conductor and detecting (using conventional equipment) the position of the diagnostic band in the spectrum. If the semi-conductor is stressed, then the position of the diagnostic band will be shifted as compared to the unstressed material with the extent of the shift being dependent on the degree of stress. Thus, by comparing the position of the diagnostic band with the position of the corresponding band in a reference condition of the semi-conductor, it is possible to determine the degree of stress in the material under investigation. In contrast to the conventional strain gauges, this technique based on Raman spectroscopy can measure deformation in 1 μm² regions by focusing a laser beam without any physical contact to the sample.

Raman microscopy has also been used to yield quantitative information on the deformation of substituted polydiacetylene, single- crystal fibres (1-3). These studies have shown that well-defined resonance Raman spectra can be obtained from single crystal fibres of substituted polydiacetylenes. When the fibres are subject to deformation, the position of the bands in the Raman spectrum are

SUBSTITUTE SHEET found to shift to lower frequency with the largest shift being for the -C=C- triple bond stretching band which moves by about 20 cm^{- 1} for 1% applied strain in single crystals. This shift is a reflection of the macroscopic deformation directly deforming the cσvalent bonds along the polymer backbone. As well as giving an important insight into molecular deformation in a polymer single crystal, the phenomenon can be employed to follow the micromechanics of deformation of fibres within a composite (3). For example, it has been possible to map the distribution of strain along a polydiacetylene fibre in a model composite consisting of a single fibre embedded in an epoxy resin bar and to determine parameters such as the critical length as a function of fibre diameter (3). Recent studies have shown this Raman strain measurement technique to be applicable to a wide variety of fibre and composite systems (4-8).

Thus, whilst it has been known in the art that Raman shift may provide a representation of strain, the existing systems are very specific and do not lend themselves to measurement of a deformation condition in a range of situations. In other words, GB-A-2 206 688 is applicable only to measuring the strain in a particular semi-conductor material whereas the polymer single crystals are relatively intractable and are also very stiff, brittle materials which would not lend themselves to a wide range of deformation measurement applications. According to the present invention there is provided a method of determining a deformation condition (as herein defined) comprising forming a tractable product into a desired configuration, said product being one which is, or is a precursor to, a polymeric material having chains including Raman active bonds which are capable of stretching but not rotation and which provide at least one diagnostic band in the Raman spectrum of a frequency dependent on the de ormation condition of the material, directing Raman excitation radiation to said material, detecting the frequency (or a repesentation thereof) of said at least one diagnostic band, comparing the value obtained with the frequency (or a representation thereof) of the corresponding band of the material, and determining said deformation condition from said comparison.

The term tractable means that the product (which is preferably a copolymer material) may be readily formed into a desired

SUBSTITUTE SHEET configuration. Thus, for example, the tractable product may be soluble so that a layer of the product may be formed from solution by evaporation of a solvent. Alternatively, the tractable product may be fusible so as to be readily formed into a desired configuration. A still further possibility is that the product is heat softenable so that it may be configured to a desired shape.

The tractable product may itself be the polymeric material with the diagnostic band in its Raman spectrum or may be converted to such a material, by curing with heat or irr diation. The polymeric material may be a homopolymer, copolymer or polymer blend (whereof at least one component is a polymer having the Raman diagnostic band).

By using tractable products it is possible relatively easily to provide polymeric materials with a Raman diagnostic band in a wide variety of configurations so that deformation conditions may be measured in a wide range of situations.

For example, the Raman active material may be readily produced in a variety of forms, eg. films, sheets and coatings. The Raman active material may be isotropic or anisotropic and have other physical properties tailored to the particular application. Thus the Raman active material may be one which is intrinsically tough and which may possess a stiffness from that characteristic of a soft rubber to that of a high-rigidity glass.

The polymeric materials used in the invention therefore overcome the limitations of the polymer single crystals which are relatively intractable in that they are insoluble and infusible, and are therefore not amenable to processing. They are also very stiff and brittle materials, and are obtainable only in limited forms such as discrete fibres, platelets or lozenges, and have properties which are highly anisotropic. As such, polymer single crystals can not readily be formed as films, sheets or coatings for use either in isolation or with other substrate materials.

Raman active materials may be isotropic or anisotropic, readily produced in a variety of forms, including films, sheets and coatings, and which are intrinsically tough, and which may possess stiffnesses controllable from those characteristic of soft rubbers to those of high-rigidity glasses. The Raman active material consist of or comprise a hard segment phase and may be a phase separated material (eg. one comprising soft and hard segment phases) with the Raman active bonds provided in one of the phases, preferably the hard segment phase.

Preferably the Raman active bonds are -C=C- or -C=C- bonds or a combination thereof. Most preferably, the material contains poly(diacetylene) units, ie. units of the repeating formula (=C-C=OC=)- In highly preferred materials for use in the invention, the poly(diacetylene) chains are cross-linked so as to form an overall network structure. Polydiacetylene chains provide resonance Raman enhancement and materials containing such chains are therefore particularly suitable for use in this invention.

In a preferred embodiment of the invention, the polymeric material including the cross-linked polydiacetylene chains is obtained by matrix polymerisation of a polymer having chains incorporating groups of the formula -C=C-C=C-. Such matrix polymer precursors may be cured by heat or irradiation to form the polydiacetylene chains. A detailed explanation of the manner in which such a matrix polymerisation proceeds to form the cross-linked polydiacetylene chains is given in the description following Example 2 (see infra) and is therefore not further detailed here.

The matrix polymer precursors are preferably block copolymers (preferably linear) with hard and soft segment phases. The -C=C-C=C- groups are preferably present in the hard segment phase. Ideally, the soft segment is totally amorphous and the hard segment has some potential for crystallisation. Such features are conducive to the matrix polymerisation reaction for the reasons detailed in the specific case detailed after Example 2 (see infra).

The matrix polymer precursors may be condensation polymers, eg. copolyurethane, a copolyurea, a copolyamide or a copolyester. Such condensation polymers may be produced from, as one reactaπt, a difunctional compound containing the -C=C-C=C- group, where the functionality refers to a group or groups capable of undergoing a condensation reaction, eg. -OH, -NH-, -COOH, -COC1. The other reactants for the condensation reaction will of course be selected depending on the polymer required. This polymer may, for example, be required to have elastomeric properties or to be more of a glass. Preferably the matrix polymer precursor is the tractable (eg. soluble) product which may be formed into a desired configuration (eg. as a surface layer on a substrate) so that subsequent curing yields the Raman active material on which the measurement of deformation condition may be made.

Examples of materials for use in the invention are copolyurethanes which have been prepared by reaction of a di-(or higher functionality) isocyanate, a monomeric, oligomeric or polymeric polyol (of functionality 2 or more), and a diacetylene diol. The reaction product comprises essentially two polymeric phases: one derived from the reaction of the isocyanate and the polyol, and the other (the urethane-diacetylene phase) resulting from the reaction between the isocyanate and the diacetylene diol. Such copolyurethane reactions may be conducted with or without added catalysts, and in either bulk (no added solvent) or in solution (with added solvent). In a bulk reaction, an exotherm may develop which effects partial cross- linking or the urethane-diacetylene phase via additional reactions in the acetylinic bonds in the diacetylene residue (ie. -C=C-C=C-), which result in the formation of polydiacetylene chains. These matrix-type reactions occurring within the urethane-diacetylene phase are referred to as cross-polymerisation which needs to be maximised by subsequent curing using heat and/or irradiation. In a solution reaction, the exotherm is reduced and cross polymerisation may not occur so that subsequent curing is essential using heat and/or irradiation to form polydiacetylene chains in the urethane-diacetylene phase. The reactions described, including in particular cross-polymerisation, will be explained more fully below with reference to specific copolyurethanes, although are also applicable mutatis mutandis to other types of polymer including diacetylene groups.

The method of the invention may be applied in a variety of ways. For example, the Raman active material (or precursor which is then converted to the Raman active material) may be applied as a layer to that area of a component (eg. of metal) which will in use be subject to deformation, as the component is subject to deformation so also is the Raman active material. The position of the diagnostic bond (or bands) in the deformed condition of the Raman active material may be measured and compared with the corresponding value(s) in the undeformed material. Thus it is possible to evaluate the degree of deformation in Raman active material and thus in the area of interest of the component. Alternatively, the Raman active material (or precursor which is then converted to the Raman active material) may be moulded or otherwise formed as a model ^■ of a particular component and the Raman spectrum is determined at a point on the model so as to obtain the position of the diagnostic band (or bands). Thus by comparison with the position of the corresponding band(s) in the spectrum of the undeformed Raman active material, it is possible to determine the degree of deformation in the model at the point of interest.

The source of Raman exciting radiation will for preference be a laser, eg. an Argon ion laser providing light at 488 and 514 nm or a Helium/Neon laser providing light at 632 nm. Alternatively, it is possible to use an IR or UV laser. It will be appreciated that the Raman spectrum of the material under investigation is obtained over the area (typically 1 μm²) on which the excitation source "impinges" on the material. This area will obviously depend on the focusing of the excitation source on the material. With a very "fine" focus it is possible to determine the deformation condition in only a very small area of the material. This renders possible a "point-by-point" analysis of the deformation condition of the material.

The Raman spectrum is obtained at the surface of the material and therefore the technique does not have any thickness limitation. In the limiting case, the Raman sensitive material could be used as a monomolecular layer, eg. a LB film.

The preceding paragraphs exemplify various applications of the method of the invention. The copolyurethanes described above are particularly suitable for use in these applications since the reactants (ie. isocyanate, polyol and diacetylene diol) may in the case of a coating, be mixed together and then applied immediately to the component (with thermal or radiation curing if necessary) so that the copolyurethane is formed in situ thereon. Alternatively, a pre-formed film coating may be used. In a further alternative the mixed reactants may be introduced into a suitable mould so that the required model is produce.

Reference is made above to measuring the position of the diagnostic band(s). It will be appreciated that this position may be measured as a frequency, wavelength, or any other convenient representation. It will also be appreciated that the deformation condition determined in the Raman active material may be an undeformed condition (in which case there will be no shift in the position of the diagnostic band(s) in the sample under investigation as compared to the reference sample) and such a result is of course of value in showing that there is no deformation at the point of investigation.

The invention will be described by way of example with reference to the following non-limiting Examples 1 to 4, and Figs. 1 to 9 of the accompanying drawings. Examples 1 and 2 illustrate the production of Raman active materials, and Examples 3 and 4 illustrate their use in measurement of a deformation condition. EXAMPLE 1

Two segmented copolyurethanes were prepared by reaction of the compounds (i) - (iii) (see Table 1), namely, pure MDI (i), a polyoxypropylene diol (ii), and a low molar mass diol chain extender (iii), namely, 2,4-hexadiyne-l,6 diol (HDD). The copolyurethanes differed in the polyoxypropylene diol (PPG) used. One of the

SUBSTITUTE SHEET copolyurethanes was based on PPG1000 for which (x + y) in formula (ii) is ca 16 and the other was based on PPG400 for which (x + y) is ca 6. Purification of Reactants and Synthesis of Urethane-Diacetylene Copolymers.

4.4'-Methylenediphenylene Diisocvanate. MDI. (ex. BDH). The aromatic MDI, in flake form, was melted and filtered through a grade 4 sinter (50 °C) using a Buchner flask and water pump (10 mm Hg). The equivalent weight of the MDI was determined to be 125 g mol-^ .

Polyoxypropylene Diols. PPG400 and PPG1000. (ex. BDH). Both diol pre-polymers were dried by vacuum rotary film evaporation (100 ^oC/0_3 mm Hg) for 5 hours. Characterisation was carried out using end-group acetylation(9) to give equivalent weights (that is, molar masses per functional group) of 203 and 498 g mol^-1 respectively, for PPG 400 and PPG1000.

2.4-Hexadiyne-l ,6 Diol. HDD. This reactant was synthesised from propargyl alcohol (ex. Fluka Chemicals) according to the procedure reported by Hay(lO), and was recrystallised from boiling toluene (111 °C) and vacuum dried (90 °C/4 h) prior to use. The pure HDD was obtained as fine white crystals and had a melting point of 112 °C (lit. 109-113 °C).

The urethane-diacetylene copolymers were formed in bulk using a one-step polymerisation process to simulate .industrial processes such as reaction injection moulding, RIM.

The rapid development of RIM has resulted in significant increases in the production of polyurethane materials, particularly segmented copolyurethane elastomers. The formation of such elastomers by RIM is by definition a one-step, bulk copolymerisation process(ll ) in which one reactant stream comprises a blend of polyether pre-polymer and diol chain extender and the other reactant stream is a liquified form of MDI. In this context, the urethane-diacetylene segmented copolymers for the present Example were formed by a hand-casting technique involving a similar one-step, bulk (rather than solution) copolymerisation process which therefore overcomes the subsequent need to remove solvent. This process is in distinct contrast to the multi-step, solution method employed by other workers(12-14) and provides an interesting comparison in terms of the structures and properties of the copolymers produced. In the bulk process, a significant rise in temperature due to the polymerisation exotherm may occur resulting in some cross-polymerisation of the diacetylene-based HS, which is not the case in the solution process. In addition, the one-step process enables HG of greater length to form although the distribution of HS lengths is broader than in a multi-step process. Consequently, in the phase-separated copolyurethane, the HS domains are less well-defined and are more easily disrupted when subjected to temperature . and deformation.

In both cases, polymerisations were carried out without catalysts using stoichiometric equivalence of -NCO and total -OH groups, and an initial reaction temperature of 80±1°C. Formulations based on the different proportions by weight of reactants used to form the two copolymers are given in Table. 1.

Typically, between 50 and 100 g of total reactants were used in formulations. The molar ratios of reactants were PPG:HDD:MDI ≡ 1 :1 :2 and the weight percentages of diacetylene-based hard segments in PPG1000/HDD/MDI and PPG4000/HDD/MDI were 22.4 and 35.5 respectively. Hard segment content is defined as the mass of HDD plus the stoichiometric equivalent of MDI divided by the total mass of the formulation.

In a typical preparation, the PPG diol and the stoichiometric equivalent amount of HDD were weighed accurately into a 250 ml sealed flanged reaction vessel, equipped with a stirrer, and immersed in a thermostatted water bath (80 ±1 °C). The pale yellow reaction mixture was stirred and degassed by applying vacuum (--0.3 mm Hg). The stoichiometric equivalent amount of molten MDI was then added to the polyol blend via a heated glass funnel. The complete reaction mixture was stirred continuously for 2 hours at 80 °C (although the initial reaction exotherm raised the temperature to ~120 °C for a few minutes). During the reaction time, the initial clear mixture became cloudy as the polymerising diacetylene-urethane hard segments and polyurethane-urethane soft segments phase-separated: there was also a colour change from pale yellow to deep orange-brown and an increase in viscosity as the overall molar mass increased. After the 2 hour period, vacuum was re-applied to de-gas the reaction mixture which was cast into picture-frame moulds, previously sprayed with silicone release agent and pre-heated to 80 °C. Polymerisation of the reaction mixtures was then completed by subjected the cast materials to a curing schedule of 120 °C for 36 hours. During curing there was a further colour change to deep purple and DSC experiments showed that no further reaction would be achieved. The cured copolyurethanes were stored in a vacuum desiccator containing silica-gel until subsequent testing.

EXAMPLE 2.

A segmented copolyurethane was prepared using the same reactants as in EXAMPLE 1 except that a second low molar mass diol, dipropylene glycol DPG (iv), was added to the formulations.

Purification of Reactants and Synthesis of the Urethane-Diacetylene Copolymer

The MDI, PPG400 and HDD were purified as described in EXAMPLE 1. The DPG (ex. Fluka Chemicals) was dried by vacuum rotary film evaporation (80 °C/0.5 mm Hg) for 10 hours, followed by vacuum distillation. Characterisation by end-group acetylation(9) gave an equivalent weight of 67 g mol-! .

A one-step, bulk polymerisation process similar to that described in EET EXAMPLE 1 was used except that immediately after adding the molten MDI, the reaction flask containing the polymerising mixture was transferred to an ice-cold water bath (to suppress the exotherm). The mixture was stirred for 2 hours and then cast into a pre-heated, picture-frame mould and cured for 6 hours at 60 °C, producing a yellowish-white and translucent, brittle glassy sheet. This "as-prepared" copolymer was soluble and GPC analysis (polystyrene calibration) gave molecular weights of Mn = 9,320 and Mw = 21 ,716 g mol^"1. Postcuring at 100 °C for 40 hours caused cross-polymerisation within the copolyurethane, causing a dramatic colour change to deep-purple and producing an insoluble and infusible, opaque material.

As in EXAMPLE 1 , the reaction was carried out without catalysts and using stoichiometric equivalence of -NCO and total -OH groups. The formulation used to form this copolyurethane is given in Table 1. Chemical Structure of Urethane-Diacetylene Copolymers

In the urethane-diacetylene copolymers, which may be described as molecular composites, the degree of reinforcement of the soft segment attained by incorporating dispersed hard segments depends on (a) hard segment content (expressed as a weight percentage, HS ), (b) the degree of phase separation and connectivity between hard a soft segments, and (c) the degree of crosslinking in the hard segment, achieved by thermal cross-polymerisation of the diacetylene moieties.

The poly(urethane-diacetylene) materials are more accurately described as segmented block copolymers of the (-AB -_n type in which the hard (A) and soft (B) segment blocks phase separate to yield a bulk material comprising a soft polymer matrix reinforced with hard glassy domains. A simplified reaction scheme showing the formation of the segmented block copolyurethanes from reactants (i) to (iii), in which -R- is the methylenediphenylene unit, is shown in Scheme 1. _

SUBSTITUTE SHEET TABLE 1. Chemical Structures and Relative Parts by Weight of Reactants in Formulations used to prepare Urethane-Diacetylene Copolymers

CH, CH,

HO-CH₂- HO-CH-CB, — O-CH, — CH-OH

(iϋ) (iv)

Ideally, block copolymers of structure (v) are linear with the soft segment being totally amorphous and the hard segment having some potential for crystallisation. The microstructural features, which define the morphology of such materials, are shown schematically ir> * ε. 1 and in the case of PPG/HDD/MDI copolymers, the thin lines represent the polyoxypropylene soft segment chains and the rectangles represent the diacetylene-MDI, hard segment chains. Phase separation results in aggregation of the hard segments into highly hydrogen-bonded, rigid

SUBSTITUTE SHEET domains. It should be noted that the size and orientation of domains varies, that there is a distribution of hard segment chain lengths and that some isolated hard segments are seen to be dispersed (or phase-mixed) within soft segment chains.

Diisocyanate + Diacetylene diol + Polyether diol

hard segment, (A) soft segment, (B)

(v) Scheme 1 . Schematic representation of the formation of segmented urethane-diacetylene copolymers

As described above, diacetylene units if ^"present in suitable aggregated form, as either crystallites or complexes, and if subjected to heat or irradiation undergo topochemical reaction to form the polydiacetylene. In the case of the segmented copolyurethanes with the type of microstructure depicted in Fig. 1 , topochemical reaction proceeds along arrays of diacetylene units attached (or absorbed) in a spatial, well-defined manner within rigid, hydrogen-bonded hard segment domains dispersed in the soft segment, polyether "solution". This type of reaction is normally referred to as matrix-polymerisation(15) and is clearly an appropriate description in the formation of the present materials since it implies that some information originally present in the template hard segment molecules and domains such as molar mass and molar mass distribution, type and degree BSTITUTE SHEET of stereoregularity, is transferred to the reaction product. Thus, assuming the hard segment block in the AB)-_n block copolymer (v) shown in Scheme 1 is part of a larger, phase separated domain, then its structure is transformed by matrix-polymerisation according to the reaction shown in Scheme 2. (This reaction scheme is analogous to that reported for the formation of substituted polydiacetylene single crystals(3,15), except that the substituents are now shown as continuing methylenediphenylene urethane chains).

The topochemical, matrix-polymerisation causes a configurational change in specific parts of the hard segments and the resulting chains forming hard segment domains are cross-polymerised. Hard segments thus comprise two different polymer chains whose axes, relative to each other, define different spatial directions. These are depicted schematically in the simplified structure (v) with the polydiacetylene (repeat structure in curved parenthesis, degree of polymerisation = m) shown as a diagonal chain with direction from

Scheme 2. Schematic representation of the topochemical. matrix-polymerisation of diacetylene units to produce cross-polvmerised hard segments in urethane-diacetylene copolymers HEET top left to bottom right, and the methylenediphenylene-based polyurethane (repeat structure in square parenthesis, degree of polymerisation = p) as a stepped, horizontal chain. The polydiacetylene chain may be regarded as a polymeric crosslink and the overall crosslink density in the hard segment is therefore very high. The overall three-dimensional structure of the finally cross-polymerised hard segments is shown by the idealised representation in Scheme 3 which depicts the domain as a completely hydrogen-bonded, ladder polymer. The degree of crosslinking depends on the extent of cross-polymerisation which for a given weight fraction of hard segments, is related to the average degree of polymerisation, p, in structure (vi) and to domain size as determined by the degree of phase separation between hard segments and the soft segment matrix.

These urethane-diacetylene copolymers, therefore, are characterised by complex molecular and morphological structures, and the various processes of copolymerisation, phase-separation and topochemical cross-polymrisation occur simultaneously to varying degrees during the materials preparation producing dramatic colour changes and intensification. This is inevitable in a one-step bulk polymerisation process in which a rapid exotherm is produced. However, the exotherm can be significantly reducedusing careful temperature control (as described in EXAMPLE 2) during copolymer preparation, which suppresses the cross-polymerisation reaction. Subsequent curing (thermal and/or irradiation) maximises cross-polymerisation and transforms the copolyurethanes into completely insoluble and infusible mterials with increased stiffness and strength. Thus, the three formulations given in Table 1 produce cross-polymerised copolyurethanes with the thermal and mechanical properties listed in Table 2.

Thus, wide ranges of stiffness, strengths, elongations and thermal properties are attainable in such materials by tailoring the chemical sturctures of the ractants and controlling the conditions of polymerisation and curing. BLE 2. Thermal and Tensile Pro erties of h -

E - Young's modulus; σ_u - tensile strength; £_u - ultimate tensile strain;

Uf - tensile work of fracture. Tg - determined by dynamic mechanical-thermal analysis.

SUBSTITUTE SHEET

Scheme 3. Idealised structures of the MDI/HDD hard segment ( a) before and (b) after cross-polymerisation, indicating the formation of a three- dimensional network. The direction of the polyurethane chains is along the z-axis, and the directions of the polydiacetylene chains are along the x- and y-axes. The filled circles represent the carbon atoms in the conjugated diacetylene and polydiacetylene chains. DEFORMATION MICROMECHANICS OF URETHANE-DIACETYLENE COPOLYMERS USING RAMAN SPECTROSCOPY

Raman Spectroscopy: Experimental

Raman spectra were obtained from cross-polymerised samples using a Raman microscope system. This is based upon a SPEX 1403 double moπochromator connected to a modified Nikon optical microscope. Spectra were obtained at a resolution of the order of ±5 cm- using the 632.8 nm line of a 10 mW He/Ne laser. A x40 objective lens with a numerical aperture of 0.65 was employed and this gave a 2 μm spot when focussed (although the objective lens was generally de-focussed to reduce the possibility of damage through excessive heating). In the case of deformed samples, the laser beam was always polarised parallel to the tensile axis.

Spectra were obtained from the surface of strips of cross-polymerised material, approximately 2 x 10 mm, during deformation using a Polymer Laboratories "Mini-mat" mechanical testing machine. This is designed specifically to fit onto the stage of an optical microscope. The strips were deformed to fixed displacements using a gauge length of the order of 30 mm. The strain was determined from the gauge length and the displacement. The load on the specimen was also monitored using a 200 N load cell. Each spectrum was determined over a period of about 10 minutes. It was found that stress relaxation took place during deformation over this period of time and the stresses quoted are mean values.

Analysis and Discussion of Raman Spectra

It was found that Raman spectroscopy could be used to obtain spectra from cross-polymerised samples of the copolymers. Fig. 2 shows a series the spectra for the PPG400/HDD/MDI glass (Fig. 2a) and the PPGIOOO/HDD/MDI elastomer (Fig. 2b). The spectra are similar to those obtained by Rubner(14) for his radiation-polymerised, polyurethane-diacetylene copolymers. The spectra in Fig. 2 contain Raman bands at 1450 cm-* and 2100 cm-* which are characteristic of the double and triple bond vibrations of the conjugated backbone of the diacetylene units(l). The variations between the Raman spectra depend upon coupling of the motion of the side group atoms with that of the backbone atoms. It can be seen that there are some slight differences in the spectra below 1400 cm-* due probably to local variations in the flexibility of side group structures. However, the two main peaks of interest at 1450 cm-** and 2100 cm-* are essentially identical.

One of the main aims of the present invention is to report the effect of deformation upon the position of the -C≡C- triple bond stretching band in the glass and elastomer. It was found that for both materials subjected to an applied tensile strain, the position of the band shifted to lower frequency as can be seen in Fig. 3a for the PPG400/HDD/DPG/MDI glass and in Fig. 3 b for the PPGIOOO/HDD/MDI elastomer. In both cases there is a significant decrease in the peak frequency of the Raman band coupled with the broadening of the band.

The effect of deformation upon the peak position is shown in more detail in Fig. 4(a) for the PPG400/HDD/DPG/MDI glass as a plot of Raman frequency versus overall applied strain durng loading (solid symbols) and unloading (closed symbols) of the copolymer. Figure 4(b) is the corresponding stress versus strain plot. In both cases, the plots are linear and the rate of Raman peak shift with overall copolymer strain is given in Table 3.

The effect of deformation upon the peak position for the PPGIOOO/HDD/MDI elastomer is shown in Fig. 5. In this case significantly higher strains could be applied because of the higher elongations possible for the elastomer. It can be seen that there is a significant decrease in the peak Raman frequence, Av, with applied strain but that the change is non-linear. On the other hand the shift in Av with applied stress shown in Fig. 5b is found to be linear within the bounds of experimental error.

As shown in Fig. 4(a), the shifts in the raman bands with deformation were found to be reversible and there was found to be a shift to higher frequency when the stress and strain were decreased. This behaviour is shown in Fig. 6 for the PPGIOOO/HDD/MDI elastomer. Another point to note from this figure is that when the elastomer is subjected to high degrees of deformation (e > 50 %) multiple peaks are obtained in the Raman spectrum. It can be seen that the band for the triple bond stretching in Fig. 6 shows several peaks when the material is deformed to a strain of 60 %. This is indicative of significant local variations in stress and strain in the hard segments. It follows a gradual broadening of the Raman band with increasing deformation and could be due to break-up of the hard segments.

TABLE 3. The Effect of Deformation upon the Peak Frequency. Av. of the 2100 cm-* Triple Bond Stretching Band for the Cross-Polymerised Urethane-Diacetylene Copolymers

(a) initial value The stress- and strain-induced Raman frequency shifts in the two cross-polymerised materials give a unique insight into the micromechanics of the deformation of the hard segments in the materials. The analysis of substituted polydiacetylene single crystals using Raman spectroscopy(3) has shown that when such crystals are deformed parallel to the chain direction the position of the bands for triple bond stretching shifts by about 20 cm^-1/ strain regardless of the crystal modulus (which is controlled by the substituent groups). Also in such materials strong Raman scattering is only obtained when the direction of the polarisation of the laser beam is parallel to the axis of the polydiacetylene molecules (ie. parallel to the fibre axis). This implies that in the materials studied in the present invention the Raman scattering comes principally from polydiacetylene units in the hard segments which are aligned parallel to the direction of polarisation of the laser beam (the stressing direction in the deformation experiments). Hence, the deformation -induced Raman frequency shifts would appear to be obtained principally from polydiacetylene units aligned parallel to the deformation axis. This means that the Raman spectroscopy essentially probes the deformation of molecules in this particular orientation. The Raman shifts measured for the materials in EXAMPLES 1 and 2 are given in Table 3.

The shifts in terms of strain are about ten times higher for the glass than for the rubber. Analysis using the series-parallel model and determination of the local strains in the different phases has shown that at a given level of overall strain the hard segment strain the PPG400/HDD/MDI glass is considerably higher than that for the hard segments in the PPGIOOO/HDD/MDI rubber. This is because the modulus of the matrix in the glass is more closely matched to that of the hard segments than in the case of the rubber. Since the Raman measurements give a direct measure of hard segment strain the higher deformation-induced shift for the glass (in terms of άAv/ά c) than for the rubber is consistent with this prediction. (The reversal of this ranking when dΔv/άσ is considered is a reflection of the higher modulus possessed by the glass).

One example of apparatus for measuring strain condition in accordance with the method of the invention is illustrated in the self-explanatory Fig. 7 of the drawings.

EXAMPLE 3

Use for the Measurement of Deformations around Defects

Materials such as the one described in EXAMPLE 2 can be employed for the measurement of the distribution of deformation around defects such as holes, notches or cracks. These are situations of particular interest in the design and performance of engineering structures. Figure 8 shows schematic diagrams of a plate with a centre hole and of a plate with a single edge notch subjected to a stress σ₀. The symbols used to define the geometry of the deformation are shown in the figure.

A plate of the PPG400/HDD/DPG/MDI material 3mm thick was prepared and cross-polymerised at 100°C for 40 hours as detailed in Example 2. It was machined into various different geometries as shown in Figure 8 and the edges were ground and polished. The different types of holes and notches used are detailed in Table 4 where a is the length of the notch, e is the radius of curvature of the notch tip or hole and w is the plate width. Raman spectra were obtained as described under "Raman Spectroscopy: Experimental". This was done on the machined plates while they were deformed in the "Mini-Mat" mechanical testing machine. The laser beam was polarised parallel to the tensile deformation axes and focussed at a spot a distance r from the edge of the hole or notch. Spectra were obtained at various levels of deformation using the 2100 cm-* Raman band. The shift of this band was determined as a function of overall applied strain, e₀, determined using a strain gauge. The strain at r, e_r, was then d(Δc)/de for the PPG400/HDD/DPG/MDI materials using Table 3. The experimental strain concentration, f_e at r, was then calculated by dividing c_r by e₀ and is given in Table 4 for the different specimen geometries. The strain concentrations were also determined theoretically (f_t) using equations given in the book of Williams(16) for the different specimen geometries and they are also listed in Table 4. It can be seen that there is good agreement between f_e and f_t.

This shows that the deformation in the material can be measured at a point in the region of stress or strain concentration ahead of a hole or notch. Measurement of the deformation at different points would allow the full mapping of deformation around such defects in plates of PPG400/HDD/DPG/MDI which could be used as models of engineering structures.

TABLE 4. Geometries and Strain Concentration Factors at the Equators of the Circular Holes and the Notch Tips

EXAMPLE 4

Use for the Measurement of Deformation in Surface Coatings

Coated specimens of the material described in EXAMPLE 2 were prepared using a solution of the as-prepared PPG400/HDD/DPG/MDI in N.N'-dimethyl acetamide. Three substrates were employed; a conventional cross-linked glassy copolyurethane, a glass fibre and a strip of aluminium. They were dipped in the solution and the coatings (0.05 mm thick) were cross-polymerised by heat treatment at 100 °C for 40 hours after solvent evaporation.

The coated specimens were then stretched under the Raman microscope using the "Mini-Mat" mechanical testing machine. The strains in the copolyurethane and the aluminium were measured using resistance strain gauges and the strain in the glass fibres was determined from the cross-head displacement. Raman spectra were obtained during deformation with the laser beam polarised parallel to the tensile axis using the 2100 cm-* Raman band. The variation in peak position of this band with specimen strain is shown for the three substrates in Figure 9. There is an approximate linear relationship between the peak position and applied strain and the slope of the lines, d(Δ»*)/d_, in Figure 9 are listed in Table 5.

TABLE 5. Strain Sensitivity of the 2100 cm-* Raman Band for the Coatings upon the Three Different Substrates.

The values of d{Δt>)/άe for the three coatings are, within experimental error, identical to that of the bulk material (PPG400/HDD/DPG/MDI in Table 3). Hence it has been demonstrated that the copolymer can be used to coat different substrates and the deformation in the substrate determined accurately from the change in positions of the 2100 cm-* band in the Raman spectrum of the coating. Although only three substrates are shown in this example, a wide variety of substrates could be employed.

REFERENCES

1. Batchelder, D.N. and Bloor, D., J.Polym.Sci.. Polvm.Phys.Ed.. 17, 569 (1979).

2. Galiotis, C, Young, R.J. and Batchelder, D.N., J.Polym.Sci.. Polvm.Phys.Ed. 21., 2483 (1983).

3. Young, R.J., Ch. 1 in "Development in Oriented Polymers - 2" (ed. I.M. Ward), Applied Science, London, 1987.

4. Day, R.J., Robinson, I.M., Zakikhani, M. and Young, R.J. Polymer. 28, 1833 (1987).

5. Robinson, I.M., Zakikhani, M., Day, R.J. and Young, R.J. J.Mater.Sci. Lett. 6, 1212 (1987).

6. Day, R.J., Piddock, V., Taylor, R., Young, R.J., and Zakikhani, M., J.Mater.Sci.. in press.

7. Young, R.J. and Day, R.J., British Polvm.J. 21, 17 (1989).

8. Young, R.J., Day, R.J., Zakikhani, M. and Robinson, I.M. Comp.Sci.Tech. 34 243 (1989).

9. Sorenson, W.R. and Campbell, T.W., "Preparative Methods of Polymer Chemistry". Interscience, New York, 1961.

10. Hay, A.S., Qrg.Chem.. 27, 3320 (1962).

11. Barksby, N., Dunn, D., Kaye, A., Stanford, J.L. and Stepto, R.F.T. in "Reaction Injection Moulding", (ed. J.E. Rresta), ACS Svmp.Ser. 270, ACS Washington DC, 83 (1985).

12. Rubner, M.F., Polvm.Mat.Sci.Eng.. 53, 683 (1985).

13. Liang, R.C. and Reiser, A., Polvm.Prepr. (AM. Chem.Soc. Div.Polvm.Chem... 26(2.. 327, (1985).

14. Rubner, M.F., Macromolecules. 19, 2**4 (1986).

15. Wegner, G., Pure & Appl.Chem.. 49, 443, (1977).

16. Williams, J.G., "Stress Analysis of Polymers", Longman, London, 1973, p.225.

STITUTE SHEET

Claims

1. A method of determining a deformation condition (as herein defined) comprising forming a tractable product into a desired configuration, said product being one which is, or is a precursor to, a polymeric material having chains including Raman active bonds which are capable of stretching but not rotation and which provide at least one diagnostic band in the Raman spectrum of a frequency dependent on the deformation condition of the material, directing Raman excitation radiation to said material, detecting the frequency (or a representation thereof) of said at least one diagnostic band, comparing the value obtained with the frequency (or representation thereof) of the corresponding band of the material in a reference condition, and determining said deformation condition from said comparison.

2. A method as claimed in claim 1 wherein the tractable product is a curable polymeric product which is formed into the desired configuration and then cured to produce said polymeric material.

3. A method as claimed in claim 1 or 2 wherein ther Raman active bonds of the polymeric material are -C=C- and/or -C=C- bonds.

4. A method as claimed in any one of claims 1 to 3 wherein the material contains poly(diacetylene) chains.

5. A method as claimed in claim 4 wherein the polydiacetylene chains are cross -linked.

6. A method as claimed in claim 5 wherein said polymeric material has been obtained by matrix polymerisation of matrix polymer precursor containing chains incorporating groups of the formula

-C=C-C≡C-.

7. A method as claimed in claim 6 wherein said matrix polymer precursor is a condensation polymer.

8. A method as claimed in claim 7 wherein the matrix polymer precursor is a copolyurethane, a copolyurea, a copolyamide or a copolyester.

9. A method as claimed in claim 8 wherein the matrix polymer precursor is a copolyurethane which is the reaction product of a di- (or higher functionality) isocyanate, a monomeric, oligomeric or polymeric polyol, and a diol containing the -C=C-C=C- group.

10. A method as claimed in claim 9 wherein the diol is HO-CH₂-C=C- C≡C-CH₂-0H.

11. A method as claimed in claim 10 or 11 wherein the isocyanate is MDI.

12. A method as claimed in any one of claims 6 to 11 wherein the matrix polymer precursor is said tractable product.

13. A method as claimed in any one of claims 6 to 12 wherein the matrix polymer precursor is cured by heat.

14. A method as claimed in any one of claims 6 to 12 wherein the matrix polymer precursor is cured by irradiation.

15. A method as claimed in any one of claims 6 to 14 wherein the Raman excitation radiation is a laser polarised parallel to the axis of the polydiacetylene chains.

16. A method as claimed in any one of claims 6 to 15 wherein the polymeric material comprises hard and soft segment phases and said polydiacetylene chains are in said hard segment phases.

17. A method of determining a deformation condition (as herein defined) comprising forming a tractable product into a desired configuration, said product being a copolymeric material with a hard segment phase including chains containing groups of the formula -C=C-C=C-, curing said product to convert said groups to poly(diacetylene) chains, directing Raman excitation radiation to said cured material, detecting the frequency (or a representation thereof) of at least one band (the diagnostic band) which is representative of the deformation condition of the material, comparing the value obtained with the frequency (or a representation thereof) of the corresponding band of the material in a reference condition, and determining said deformation condition from said comparison.

SUBSTITUTE SHEET