Aging of a Poly(vinyl acetate)-Based White Glue and Its Durability in Contemporary Artworks (2024)

Massimo Lazzari, Conceptualization, Formal analysis, Investigation, Writing – original draft, Writing – review & editing, Project administration, Funding acquisition^1,2,* and Thais López-Morán, Conceptualization, Data curation³

2.1. Materials

The serial artwork Palette (from the portfolio For Joseph Beuys, edition of 90) was realized by Tony Cragg in 1986. It is made of a thick layer of small plastic granulates glued onto a wood support as a painter’s palette, with overall dimensions of 74 × 57 × 1.5 cm (Figure S1 in the Supporting Information). Analyses were carried out on 9 plastic granulates and 3 adhesive fragments, with a dimension of a few cubic millimeters, detached from the surface and gathered in the artwork case (Figure S2).

Tierra, ladrillo y agua I, II, III e IV, realized in 2001 by Dario Villalba, consists of bricks, gravel, soil, sand, and dry vegetal residues glued on 4 canvases, with dimensions 200 × 250 cm (Figure S3). Detached debris of different materials, including 4 whitish adhesive fragments (Figure S4), were used for identification analysis. The inorganic and vegetable debris were not the subject of this study and will not be discussed further.

The same commercial white glue used by Villalba, named cola blanca uso general (Rayt, Badalona, Spain), was bought in a local store and used as a model PVAC-based adhesive.

2.2. Accelerated Aging and Characterization Techniques

Films for aging treatments, with a thickness of 80–100 μm, were prepared by spreading the commercial emulsion onto glass slides for all the measurements except for transmission FTIR spectroscopy measurements where KBr disks were used as support. Samples were dried at room temperature for 48 h and then 24 h at 80 °C, until reaching a constant weight.

Accelerated photoaging treatment was performed in a high-speed exposure unit Suntest CPS+ (Heraeus), equipped with a xenon light source with constant irradiation at 765 W/m². A glass filter with a cut-off at λ < 295 nm was used to exclude radiation more energetic than that of outdoor daylight exposure, in order to reproduce solar irradiation. The maximum temperature of the samples during irradiation was 44 °C black panel temperature, as maintained by a forced-air circulation system. Isothermal treatments were performed at constant temperatures in the range of 80–150 °C in a forced-air circulation oven.

FTIR spectra were collected using a Thermo Nicolet 6700 FTIR instrument equipped with a Smart Endurance device for ATR measurements and a liquid nitrogen-cooled mercury cadmium telluride (MCT) detector (spectral range 4000–670 cm⁻¹) at 4 cm⁻¹ resolution for 128 scans. Data were processed with Omnic 8.1 by Thermo Nicolet. Samples from artworks were analyzed as received. Color measurements were performed with a portable Konica Minolta CM 700d spectrophotometer (Konica Minolta, INC., Tokyo, Japan). Color space was the CIELAB with D65 standard illuminant and 10° observer, ø 6 mm illumination area. The International Commission on Illumination (CIE) developed this L*a*b* color model in 1976 to measure objective color and calculate color differences. DSC thermograms were obtained with a Q200 (TA Instruments, New Castle, DE, United States) calorimeter equipped with a refrigerated cooling system in the temperature range from −70 to 100 °C, using 10–15 mg samples with a scanning rate of 20 °C min⁻¹, under a 50 mL min⁻¹ nitrogen flow. Weight changes in supported films and insoluble fractions were determined gravimetrically with an analytical balance and Gr-200 with a 0.0001 g repeatability and ±0.0002 g linearity. Surface-enhanced Raman spectroscopy (SERS) spectra were recorded with a Renishaw InVia Flex Raman spectrometer (Wotton-under-Edge, UK) equipped with a continuous wave laser with emission at 514 nm with gratings of 1800 lines mm⁻¹. Measurements were performed with a long 0.50 NA NPlan long working distance objective Leica 566036 (Leica Microsystems Gmbh, Wetzlar, Germany) operating with a 65 µm slit opening, with a 10 s accumulation and 1% of the nominal power, corresponding to 0.09 mW. The spectrometer slit opening was 65 µm, using the detector Renishaw CCD 576 × 400 pix. Sampling from aged surfaces was carried out by 0.5 × 0.5 cm silicone strip samplers fabricated by casting and thermal curing of a liquid prepolymer (Silgard 184, Dow Corning, Midland, MI, United States) under vacuum at 80 °C for 2 h. Physisorbed molecules were then dissolved in around 10 µL of chloroform and transferred onto SERS-active Al-coated 3D nanostructures. More details on the sampling procedure and the fabrication of the SERS substrates were reported elsewhere [20,21].

3.1. Materials Characterization

A total of 9 plastic granulates and 3 adhesive fragments released from the artwork Palette were analyzed by ATR-FTIR spectroscopy to identify the polymeric components and enlighten whether aging may justify their detachment from the support. As typical in Cragg’s work [22], the granulates consisted of common industrial polymers, such as polyethylene (PE), polypropylene (PP), polystyrene, and acrylonitrile-butadiene-styrene copolymer (ABS) (examples in Figure S2). They did not show any relevant sign of degradation but were always coated with a thin layer of adhesive on the surface. Common features are marked in the examples of ATR-FTIR spectra shown in Figure 1a–c. In addition, the spectra of fragments of the adhesive (example in Figure 1d) allowed the identification of PVAC as its main component. The main peaks at 3000–2800 cm⁻¹ (CH₂ and CH₃ stretching), 1735 cm⁻¹ (C=O stretching), 1434 and 1372 cm⁻¹ (CH₂ and CH₃ bending), and 1023 cm⁻¹ (C-O stretching) are due to PVAC. The broad –OH stretching absorption in the 3600–3100 cm⁻¹ is associated with poly(vinyl alcohol) [23], a typical secondary component of PVAC-based white glues.

Also, in the case of Villalba’s artwork, the ATR-FTIR spectrum of the whitish debris (Figure 1e) shows the series of peaks detailed for Cragg’s adhesive, with only some difference in the relative intensity of some minor peaks, which allowed to identify the adhesive used for its assembly as a PVAC-based glue. In particular, with respect to Cragg’s glue residue, the spectrum in Figure 1e shows a more complex OH absorption band and other minor peaks, e.g., at approximately 1620 cm⁻¹, indicating additional glue components. In addition, the higher relative intensity of the peak at around 1100 cm⁻¹ with respect to the PVAC C-O stretching peak at 1023 cm⁻¹ in Figure 1e in comparison with Figure 1d can be related to a different amount of plasticizers [23].

PVAC emulsions used as adhesives essentially consist of a PVAC hom*opolymer produced in the presence of a polymeric protective colloid, usually low molecular weight PVOH, and other additives, such as plasticizers and coalescence agents [24,25]. Despite compositional differences within this class of adhesives, as recently studied by De Sá et al. [23], long-term aging is expected to follow common features. In this work, we considered the behavior of the commercial white glue stated by Villalba during a private conversation, as representative of the entire class.

The technical sheet of the Rayt commercial white glue only indicates the solid fraction (approximately 50 wt.%), PVAC as the main component, PVOH as a minor component, gypsum (CaSO₄·2H₂O), and other unidentified additives. Elemental analysis, which detected a high amount of sulfur, and TGA allowed to confirm the presence of gypsum and its quantification (19 wt.%), while GC/MS unveiled the presence of dimethyl succinate and dibutyl phthalate (DBP).

3.2. Accelerated and Natural Aging

To predict the long-term degradative behavior of the critical adhesive material in the artworks and better understand its loss of cohesive strength, dry films of the commercial glue were exposed to simulated aging conditions [26]. As a common practice in polymer science, all the processes occurring during natural aging may be artificially accelerated through either (a) irradiation in a photoaging device equipped with a solar lamp or (b) isothermal treatment at a temperature higher than the environmental one, not triggering unexpected reactions in both cases. Under such limitations, accelerated aging treatments are only expected to induce the same chemical and physical changes, including oxidation and additives’ migration, at a higher rate than those occurring under museum conditions [22].

Reference samples in the form of films were prepared by water evaporation onto a solid substrate, either glass or quartz, and dried until constant weight. Differently from the case of pure PVAC hom*opolymer, our dried films showed a very limited solubility in common organic solvents such as chloroform and acetone, thus suggesting an emulsion formulation also containing small amounts of a cross-linker. In addition, by comparison with the adhesive fragments sampled from the artworks (Figure 1d,e), the transmission FTIR spectrum of the dried reference adhesive (Figure 1f) shows the presence of two small peaks at approximately 1580 and 1540 cm⁻¹ due to the plasticizer, i.e., DBP identified by GC/MS.

Films submitted to photoaging showed negligible weight loss for treatment up to 1500 h (<1%), whereas the surface turned from transparent to white already after 500 h due to the so-called blooming. Both phenomena may be related to the partial migration of the plasticizer, i.e., DBP, to the surface, and then partially evaporating. As a further confirmation of the fact that a complete release of DBP from a PVAC matrix has been observed only after 2000 h aging in the same type of photodegradation device [6], transmission FTIR spectra of 1500 h aged films only showed minor modification in the range 1600–1500 cm⁻¹, which is difficult to be quantified.

Due to the high stability of PVAC-based materials [6,10,12], we also carried out simulated aging at isothermal conditions, at temperatures high enough to accelerate the degradation processes at a higher rate than photoaging in solar lamp devices. In addition, previous studies by Allen and co-workers did not show changes at 120 °C [27]. On the other hand, our isothermal treatment essays at higher temperatures entailed the same changes observed at 130 °C as detailed below, but in a much shorter aging time.

The weight loss during the isothermal treatment at 130 °C was more marked than during the photoaging and increased over time, up to a final 5% loss after 550 h (Figure 2). Even if, in principle, the partial decomposition of gypsum with crystallization water release cannot be excluded, achieving the same plateau value also at higher constant temperature treatments revealed the release of some other molecules. The corresponding chemical changes were monitored by FTIR spectroscopy, showing small spectral modifications very similar for the treatments at 130 °C and higher temperatures (selected spectra are shown in Figure 3). They essentially consist of a decrease in the relative intensity of some peaks or shoulders, which may be related to the volatilization of the plasticizer. In particular, the decrease in the 1580 and 1542 cm⁻¹ peaks until their virtual disappearance, together with the decrease in shoulders/secondary peaks at approximately 1460, 1120, and 1075 cm⁻¹. No new peaks were formed during the isothermal treatment for up to 550 h, confirming the good oxidative stability of PVAC [28], nor were there signs of the well-known process of deacetylation that would lead to the formation of C=C double bonds [10]. Moreover, even the use of a technique more sensitive than FTIR spectroscopy to detect the formation of new groups, such as nuclear magnetic resonance, forcedly at the solid state, did not offer further information.

At the same time, the visual inspection of the aged films showed an extensive color change since the beginning of the isothermal treatment (Figure 2), quantified using a spectrophotometer in the CIELAB color space (Table S1). ΔE reaches a high value from 24 h up to approximately 45 at 550 h. A more detailed analysis of the evolution of L*, a*, and b* coordinates indicates that the most relevant variations are those of ΔL* and Δb*, indicating a decrease in luminosity and a progressive yellowing, respectively. The formation of chromophores is considered an early indicator of degradation processes [22]. Although not confirmed, the progressive darkening is due to the formation of increasingly longer sequences of C=C double bonds over time, although in an amount not detectable by FTIR spectroscopy.

The evolution of the film’s aging may additionally be followed through an indirect evaluation of the changes influencing molecular motion by DSC. The temperature of glass transition, T_g, of PVAC hom*opolymer is in the range of 35–40 °C, depending on the molecular weight [29], but it is expected to be much lower in dried white glue due to the presence of PVOH and other low molecular weight additives, especially plasticizers. During the isothermal treatment at 130 °C, the T_g increases from an initial value of 7 °C to 35 °C after 550 h aging (Table 1, with examples of thermograms in Figure 4). This behavior may directly be related to the release of the plasticizer, thus confirming FTIR spectroscopy highlights. Films submitted to photoaging only showed negligible T_g changes.

A direct indication of the release of DBP was obtained through a surface analysis strategy based on SERS measurements, which was developed to identify degradation markers with low molecular weight from polymer surfaces [21] and further optimized to detect small molecules of dyes and pigments from prints. Films were sampled using silicone strip samplers pressed onto the exposed film surface for 30 s. The low molecular weight molecules eventually physisorbed by the sampler were then dissolved in chloroform and transferred onto Al-coated nanostructured SERS-active substrates. Further details on the substrates, with an enhancement factor of 10⁹ at the excitation wavelength of election, i.e., 514 nm, their fabrication, and use are reported elsewhere [20]. Dried films before aging did not reveal the presence of any molecules on their surface, whereas SERS spectra of surface molecules from aged films after at least 100 h photoaging or 24 h isothermal treatment at 130 °C showed spectra like that shown in Figure 5. Even if an unambiguous identification is impossible, characteristic peaks at approximately 1745, 1602, and 1060 cm⁻¹ are compatible with phthalates [30]. This hypothesis agrees with the features highlighted by FTIR spectroscopy and DSC, justified with a plasticizer loss without showing any structural changes associated with PVAC degradation.

Validation of the results from monitoring the accelerated aging of the commercial white glue was carried out via direct comparison with its natural aging under milder museum conditions for more than 20 or 30 years in the two selected artworks. As it was already clear that FTIR spectra of the adhesive used in artworks did not contain appreciable amounts of plasticizers, i.e., no peaks in 1600–1500 cm⁻¹ in the spectra in Figure 1d,e, such adhesive fragments were also submitted to DSC measurements to compare their molecular mobility with that of the “as applied” glue, immediately after drying. In both cases, the thermograms showed T_g higher than the initial, and precisely 16 °C and 30 °C for the Villalba and Cragg’s adhesive, respectively (example in Figure S5).

PVAC tends to creep under a sustained load [31], and adding a plasticizer in its formulation has, among others, the effect of lowering hardness and strength and increasing creep [2]. Thus, increasing adhesion is a fundamental characteristic to guarantee the stability of the works of art, especially in the case of Villalba’s work, based on the assembly of heavier objects. The observed loss of plasticizer, consequent T_g and hardness increase, and partial loss of creep strength during aging fit perfectly with the detachment of debris with different weights from the artworks. Concerning their visual appearance, it is also worth noting that artwork fragments also showed extensive surface whitening, as seen during the aging simulation. In addition, contrary to what was observed during isothermal treatments, they did not show any yellowing. This behavior may be due to the fact that longer times of natural aging are necessary to accumulate a concentration of chromophores, tentatively sequences of C=C double bonds, high enough to be visible.

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

1. Saunders K.J. Organic Polymer Chemistry: An Introduction to the Organic Chemistry of Adhesives, Fibres, Paints, Plastics and Rubbers. Springer; Dordrecht, The Netherlands: 1988. Poly(Vinyl Acetate) and Related Polymers; pp. 113–124. [Google Scholar]

2. Conner A. Wood: Adhesives. Encycl. Mater. Sci. Technol. 2001:9583–9599. doi:10.1016/B0-08-043152-6/01734-4. [CrossRef] [Google Scholar]

3. Novak M., Ormsby B. Poly(Vinyl Acetate) Paints: A Literature Review of Material Properties, Ageing Characteristics, and Conservation Challenges. Polymers. 2023;15:4348. doi:10.3390/polym15224348. [PMC free article] [PubMed] [CrossRef] [Google Scholar]

4. Grassie N. The Thermal Degradation of Polyvinyl Acetate: Part 2.—Determination of the Rate Constants of the Primary Processes Involved in the Elimination of Acetic Acid. Trans. Faraday Soc. 1953;49:835–842. doi:10.1039/TF9534900835. [CrossRef] [Google Scholar]

5. Buchanan K.J., McGill W.J. Photodegradation of Poly(Vinyl Esters)-I. Formation and Quantitative Measurement of Volatile Products. Eur. Polym. J. 1980;16:309–312. doi:10.1016/0014-3057(80)90074-9. [CrossRef] [Google Scholar]

6. Toja F., Saviello D., Nevin A., Comelli D., Lazzari M., Levi M., Toniolo L. The Degradation of Poly(Vinyl Acetate) as a Material for Design Objects: A Multi-Analytical Study of the Effect of Dibutyl Phthalate Plasticizer. Part 1. Polym. Degrad. Stab. 2012;97:2441–2448. doi:10.1016/j.polymdegradstab.2012.07.018. [CrossRef] [Google Scholar]

7. Wei S., Pintus V., Schreiner M. Photochemical Degradation Study of Polyvinyl Acetate Paints Used in Artworks by Py-GC/MS. J. Anal. Appl. Pyrolysis. 2012;97:158–163. doi:10.1016/j.jaap.2012.05.004. [PMC free article] [PubMed] [CrossRef] [Google Scholar]

8. Pintus V., Wei S., Schreiner M. Accelerated UV Ageing Studies of Acrylic, Alkyd, and Polyvinyl Acetate Paints: Influence of Inorganic Pigments. Microchem. J. 2016;124:949–961. doi:10.1016/j.microc.2015.07.009. [CrossRef] [Google Scholar]

9. Doménech-Carbó M.T., Silva M.F., Aura-Castro E., Fuster-López L., Kröner S., Martínez-Bazán M.L., Más-Barberá X., Mecklenburg M.F., Osete-Cortina L., Doménech A., et al. Analytical and Bioanalytical Chemistry. Volume 399. Springer; Berlin/Heidelberg, Germany: 2011. Study of Behaviour on Simulated Daylight Ageing of Artists’ Acrylic and Poly(Vinyl Acetate) Paint Films; pp. 2921–2937. [PubMed] [Google Scholar]

10. Ferreira J.L., Melo M.J., Ramos A.M. Poly(Vinyl Acetate) Paints in Works of Art: A Photochemical Approach. Part 1. Polym. Degrad. Stab. 2010;95:453–461. doi:10.1016/j.polymdegradstab.2010.01.015. [CrossRef] [Google Scholar]

11. Toja F., Saviello D., Nevin A., Comelli D., Lazzari M., Valentini G., Toniolo L. The Degradation of Poly(Vinyl Acetate) as a Material for Design Objects: A Multi-Analytical Study of the Cocoon Lamps. Part 2. Polym. Degrad. Stab. 2013;98:2215–2223. doi:10.1016/j.polymdegradstab.2013.08.021. [CrossRef] [Google Scholar]

12. Chelazzi D., Chevalier A., Pizzorusso G., Giorgi R., Menu M., Baglioni P. Characterization and Degradation of Poly(Vinyl Acetate)-Based Adhesives for Canvas Paintings. Polym. Degrad. Stab. 2014;107:314–320. doi:10.1016/j.polymdegradstab.2013.12.028. [CrossRef] [Google Scholar]

13. Ploeger R., René De La Rie E., McGlinchey C.W., Palmer M., Maines C.A., Chiantore O. The Long-Term Stability of a Popular Heat-Seal Adhesive for the Conservation of Painted Cultural Objects. Polym. Degrad. Stab. 2014;107:307–313. doi:10.1016/j.polymdegradstab.2014.01.031. [CrossRef] [Google Scholar]

14. Gómez-Sánchez E., Agnini E., Bridarolli A., Rolland R., Kunz S., Röhrs S., Harrivelle C., Casanova E., Simon S., Calparsoro-Forcada E. Evaluation of the Solubility and Colour Changes of Artificially and Naturally Aged Adhesives for the Conservation of Ceramics and Glass. Stud. Conserv. 2022;67:500–517. doi:10.1080/00393630.2021.1995262. [CrossRef] [Google Scholar]

15. Centenaro S., Franceschin G., Cattaruzza E., Traviglia A. Consolidation and Coating Treatments for Glass in the Cultural Heritage Field: A Review. J. Cult. Herit. 2023;64:132–143. doi:10.1016/j.culher.2023.09.006. [CrossRef] [Google Scholar]

16. Melo R.H.R.Q., Falcão J.R., Bersch J.D., Baptista D.T., Masuero A.B. Performance and Durability of Paints for the Conservation of Historic Façades. Buildings. 2024;14:1016. doi:10.3390/buildings14041016. [CrossRef] [Google Scholar]

17. Chiozza F., Santoni I., Pizzo B. Discoloration of Poly(Vinyl Acetate) (PVAc) Gluelines in Wood Assemblies. Polym. Degrad. Stab. 2018;157:90–99. doi:10.1016/j.polymdegradstab.2018.10.003. [CrossRef] [Google Scholar]

18. Down J.L., MacDonald M.A., Tétreault J., Williams S. Adhesive Testing at the Canadian Conservation Institute—An Evaluation of Selected Poly(Vinyl Acetate) and Acrylic Adhesives. Stud. Conserv. 1996;41:19–44. doi:10.1179/sic.1996.41.1.19. [CrossRef] [Google Scholar]

19. Down J.L. The Evaluation of Selected Poly(Vinyl Acetate) and Acrylic Adhesives: A Final Research Update. Stud. Conserv. 2015;60:33–54. doi:10.1179/2047058414Y.0000000129. [CrossRef] [Google Scholar]

20. Gómez M., Kadkhodazadeh S., Lazzari M. Surface Enhanced Raman Scattering (SERS) in the Visible Range on Scalable Aluminum-Coated Platforms. Chem. Commun. 2018;54:10638–10641. doi:10.1039/C8CC04280B. [PubMed] [CrossRef] [Google Scholar]

21. Gómez M., Reggio D., Lazzari M. Detection of Degradation Markers from Polymer Surfaces by a Novel SERS-Based Strategy. Talanta. 2019;191:156–161. doi:10.1016/j.talanta.2018.08.046. [PubMed] [CrossRef] [Google Scholar]

22. Lazzari M., Reggio D. What Fate for Plastics in Artworks? An Overview of Their Identification and Degradative Behaviour. Polymers. 2021;13:883. doi:10.3390/polym13060883. [PMC free article] [PubMed] [CrossRef] [Google Scholar]

23. De Sá S.F., Viana C., Ferreira J.L. Tracing Poly(Vinyl Acetate) Emulsions by Infrared and Raman Spectroscopies: Identification of Spectral Markers. Polymers. 2021;13:3609. doi:10.3390/polym13213609. [PMC free article] [PubMed] [CrossRef] [Google Scholar]

24. Silva M.F., Doménech-Carbó M.T., Fuster-Lopéz L., Martín-Rey S., Mecklenburg M.F. Determination of the Plasticizer Content in Poly(Vinyl Acetate) Paint Medium by Pyrolysis-Silylation-Gas Chromatography-Mass Spectrometry. J. Anal. Appl. Pyrolysis. 2009;85:487–491. doi:10.1016/j.jaap.2008.11.020. [CrossRef] [Google Scholar]

25. Silva M.F., Doménech-Carbó M.T., Osete-Cortina L. Characterization of Additives of PVAc and Acrylic Waterborne Dispersions and Paints by Analytical Pyrolysis-GC-MS and Pyrolysis-Silylation-GC-MS. J. Anal. Appl. Pyrolysis. 2015;113:606–620. doi:10.1016/j.jaap.2015.04.011. [CrossRef] [Google Scholar]

26. Clough R., Billingham N., Gillen K. Polymer Durability: Degradation, Stabilization, and Lifetime Prediction. American Chemical Society; Washington, DC, USA: 1996. (ACS Advances in Chemistry Series 249). [Google Scholar]

27. Rodríguez-Vázquez M., Liauw C.M., Allen N.S., Edge M., Fontan E. Degradation and Stabilisation of Poly(Ethylene-Stat-Vinyl Acetate): 1—Spectroscopic and Rheological Examination of Thermal and Thermo-Oxidative Degradation Mechanisms. Polym. Degrad. Stab. 2006;91:154–164. doi:10.1016/j.polymdegradstab.2005.04.034. [CrossRef] [Google Scholar]

28. Pereira A.I.M.L. The Perfect Paint in Modern Art Conservation: A Comparative Study of 21st Century Vinyl Emulsions. NOVA School of Science and Technology; Lisbon, Portugal: 2015. [Google Scholar]

29. Brandrup J., Immergut E.H., Grulke E.A., Abe A., Bloch D.R., editors. Polymer Handbook. 4th ed. John Wiley & Sons, Inc.; Hoboken, NJ, USA: 1999. [Google Scholar]

30. Nørbygaard T., Berg R.W. Analysis of Phthalate Ester Content in Poly(Vinyl Chloride) Plastics by Means of Fourier Transform Raman Spectroscopy. Appl. Spectrosc. 2004;58:410–413. doi:10.1366/000370204773580248. [PubMed] [CrossRef] [Google Scholar]

31. Hilles H., Monroy F. Dilational Creep Compliance in Langmuir Polymer Films. Soft Matter. 2011;7:7790–7796. doi:10.1039/c1sm05255a. [CrossRef] [Google Scholar]