JAIC 1991, Volume 30, Number 2, Article 8 (pp. 203 to 213)
JAIC online
Journal of the American Institute for Conservation
JAIC 1991, Volume 30, Number 2, Article 8 (pp. 203 to 213)





FOUR MOLECULAR weight PVAC resins were used: AYAT (MW = 226,000), AYAF (MW = 169,000), AYAA (MW = 110,000), and AYAC (MW = 14,500). (The molecular weights reported are the weight average molecular weights of PVAC grades as determined by gel permeation chromatography [GPC] [Druzik 1987]). AYAT and AYAF resins were used to cast films from different solvent systems. AYAA, AYAC, and AYAT resin samples were analyzed by FTIR spectroscopy.

Solvents for solution preparation included chloroform, toluene, acetone, and a solution of 10 parts by volume acetone, 10 parts by volume ethanol, and 1 part by volume water.


Four percent solutions (w[g]/v[ml]) of AYAT in each of the four solvent systems were cast upon a Teflon FEP (polyhexafluropropylene) surface in break-apart molds as previously described (Hansen and Taketomo 1989). Two series of films with an area of 0.0225 m2 were cast: one series 0.08 mm thick and dried for 50 days and one 0.03 mm thick and dried for 180 days. A constant temperature of 22C and RH of 65% were maintained throughout the drying intervals.

Additionally, 4% solutions of AYAF in acetone and chloroform were cast in the same manner. Films 0.03 mm thick were dried under the same conditions for 180 days.


The stress-to-break, strain-to-break, yield-stress, and initial modulus for each drying interval were determined for films of AYAT and AYAF by tensile fracture at constant elongation according to the specifications in ASTM D-2370–82(87), “Standard Test Method of Tensile Properties of Organic Coatings” (ASTM 1989). Values were electronically recorded from an Instron 4201 Universal Testing Instrument and analyzed with the Instron software Series IX Automated Materials Testing System. The temperature and humidity were maintained at 20C and 65% RH, respectively. Strips 12 mm in width were precision cut using a Thwing Albert JDC Precision Sample Cutter and were fractured at a cross-head speed of 50 mm/min from an initial gauge length of 76 mm. Film thickness over the gauge length varied less than 5%, determined with an eddy current–type coating thickness gauge which can measure the thickness of a dielectric film (Vector TX1). The tensile values reported for a cast film are the mean of four to nine samples, depending upon the range of the values obtained on the sample sets.


Films of AYAT were analyzed for residual amounts of solvent. Films at the 50-day drying interval were analyzed for weight percentage of residual solvent at an analytical service center (Baxter Analytical Labs, Irvine, Calif.) by direct insertion probe thermal desorption mass spectrometry using a HP 5987A Mass Spectrometer System. Solvent content was estimated from area ratios of the thermal desorption profile. Direct insertion probe conditions were a temperature ramp from 30C to 350C at 15C/min with a hold at 350C for 5 minutes.

After drying for 180 days films were analyzed on an HP 5890 Gas Chromatograph (GC) with a CDC 120 pyroprobe. The samples were pyrolyzed at 200C with the ramp turned off and an interface temperature of 40C. Optimal separation was obtained on a 25 m Carbowax 20M column with an initial temperature of 100C ramped to 150C at 5C/min. Helium carrier gas was used at a flow rate of 30 ml/min. The FID detector was held at 220C. Internal standards were used to identify and calibrate the solvent peaks. In addition, reference films were prepared for comparison in which the amount of solvent was determined by a gravimetric method of analysis of films during the drying process. Due to the limited solubility of PVAC AYAT in pure ethanol, this method was not used to determine residual amounts of ethanol.


The Tg data of bulk AYAT and AYAF cast from the four solvent systems were obtained by differential scanning calorimetry (DSC). The method used to obtain the Tg has been previously described (Schilling 1989).


FTIR spectra were obtained at 4 cm−1 resolution on a DIGILAB 15-E FTIR spectrophotometer equipped with a Motorola 3200 computer and a dry nitrogen purge. A wide-range, cryogenically cooled MCT detector was used to examine the mid-IR region from 4000–500 cm−1. Each spectra was the accumulation of 200 scans.

Sample preparation for the infrared analysis of the bulk resin consisted of dissolving the resins (AYAA, AYAT, AYAC) in chloroform or acetone and depositing the solution on powdered KBr. The solvent was allowed to evaporate and then the mixture was pressed into a transparent pellet under vacuum. Additionally, for AYAT, a portion of bulk resin was powdered and mixed with KBr. The pellets were analyzed in transmissive mode using a 3 mm diameter beam.


The physical properties of solution-cast films of thermoplastic polymers are influenced not only by the solvent type and solvent polarity, but also by:

  • the concentration of the solution used for casting;
  • the degree of plasticization of the polymer film by retained solvent;
  • the thickness of the cast film;
  • the substrate used in casting;
  • the thermal history; and,
  • for polymers that incorporate water, both the RH maintained during the drying process and the hygroscopicity of the solvent.

All of these factors were considered in the choice of the film history of the samples prepared for testing. The objective was to isolate as much as possible the results of the influence of one parameter—the solvent chosen for casting.

The solvents were chosen for their anticipated effect on the final physical properties of the cast films. All four solvents (chloroform, acetone, toluene, and ethanol) are good solvents for PVAC in that the molecular conformation is extended in comparison to the conformation in a theta solvent, where the polymer chain is neither extended nor contracted by interaction with the solvent. The best solvent in this series is chloroform because, as can be shown by both the Huggins coefficient and the intrinsic viscosity, the polymer chain is most extended (Olayemi and Adeyeye 1982). Acetone, in addition to being a relatively poorer solvent for PVAC than chloroform, has a greater degree of polarity. The addition of ethanol to acetone increases the polarity of the solution in comparison to pure acetone. Toluene was included as a solvent due to the well-known long-range solvent retention (Newman and Nunn 1975), which allowed the effects of plasticization by residual solvent to be assessed and compared with the effects of solvent type.

Relatively dilute (4% w[g]/v[ml]) solutions were used, because higher concentrations of polymer may promote interchain interactions, and thus the physical properties may show a dependence on the concentration. This effect has been shown for the refractive index of some polymers (Ashkok and Avadhani 1987). Therefore, a consistent concentration was maintained throughout the experiment.

Although thicker films are more convenient for the physical testing of the released film, the original thickness of the films (0.08 mm) tested at a 50-day interval contained higher concentrations of both chloroform and toluene than was desirable (table 1). Because solvent retention is dependent upon both the square root of the film thickness and the amount of time following casting (Newman et al. 1975), the amount of retained solvent was reduced by decreasing the thickness of the films and increasing the drying time to six months.

TABLE 1 Solvent Retention and Glass Transition Temperature of AYAT Films at 50 Days and 180 Days after Casting from Different Solvents

The casting substrate, Teflon, permits easy release of a film of PVAC at any interval following casting. The films can therefore be studied meaningfully in relation to coatings, since the system resembles a coated surface that allows solvent release from one surface only. In addition, the surface properties of films cast on Teflon should be isotropic at both the air and Teflon interfaces because Teflon is a low surface energy substrate. Anisotropic surface properties have been observed for poly(methyl methacrylate) films formed on metal substrates (Carre et al. 1980).

It was also decided to study the films without annealing the resin since most thermoplastic polymers are applied at room temperature to objects that can not be heated above the glass transition temperature in order to drive off residual solvent. In most laboratory studies of basic “properties” resins are first annealed because the thermal history of the polymer is known to affect certain properties, especially the Tg(Slade and Jenkins 1970).

A constant humidity level was also maintained during the drying intervals to minimize variation due to the water content of the films.

Copyright 1991 American Institute for Conservation of Historic and Artistic Works