What makes your shortening suitable for fancy croissants, puff and Danish pastry?

By Braulio A. Macias-Rodriguez and Alejandro G. Marangoni

In This Section

January 2018

  • Roll-in shortenings are firm yet malleable specialty fats used in the manufacture of laminated doughs.
  • Their functionality or mechanical performance is largely determined by their bulk rheological properties at large deformations.
  • Large oscillatory shear tests reveal that roll-in shortenings are more effective at dissipating energy than other bakery fats, a property that explains their ability to bear stresses during lamination.

Shortenings are fats used for many applications, including cakes, icings, and laminated doughs, among others. As indicated by their name, their main role is to “shorten” baked products by preventing interaction of gluten and starch particles. Other important functions include structuring, lubricating, foaming, emulsifying, and so on. According to the functional requirements, shortenings fall into various categories including those intended for roll-in applications.

A roll-in shortening, used in laminated doughs, serves as barrier by forming continuous thin fat films that prevent fusion of dough layers. To do so, this roll-in material must “survive” deformations imparted during co-extrusion, sheeting, and folding of dough. If the shortening is too soft, it will be absorbed into the dough or squeezed out; if it is too firm and brittle, it will rupture the dough. Either case yields the same disastrous result: small pastries with poor crumb structure, low lift and volume.

To achieve good roll-in functionality, formulation and crystallization conditions are adequately customized to meet a set of physical criteria. Roll-in shortenings are formulated with higher contents of tri-saturated, and tri-unsaturated triacylglycerols (TAGs) consisting of trans fatty acids in some cases. This represents a major shortcoming as the industry aims to rid fat products of industrial trans fats to comply with the current regulatory framework.

Roll-in shortenings are crystallized under substantial cooling and shear conditions using a scraped-surface heat exchanger, worker units, and extrusion passages, which contribute to breakage of crystal aggregates, and their orientation into layer-like microstructures. Some of the physical parameters that are believed to describe roll-in functionality include higher solid fat content and melting points, ß’ polymorphism, and high yield values. Nevertheless, it was recently found that bakery shortenings may share similar physical properties. For example, an icing shortening may share a similar melting point, solid fat content (though wider melting ranges), and polymorphic behavior with a roll-in shortening. Likewise, viscoelastic attributes, such as yield values, could not be correlated with the usability of shortenings. These findings were particularly important (yet puzzling) since these properties had been long used to design for baking functionality.

Rheologically speaking, a glimpse into the differences among shortenings may be gained partly by tactile perception (using the so-called “thumb or finger” test) and partly by visual observation during deformation—still common industry practices. A roll-in shortening may be perceived as firm yet malleable (“shape” retaining) during handling in contrast to an all-purpose shortening that would be perceived as a firm yet brittle material. Indeed, during compression a roll-in shortening does not seem to develop visible cracks, whereas an all-purpose shortening does (Fig. 1). These characteristics are important factors in determining the quality of lamination, and whether the dough bakes into a puff and flaky pastry.

Figure 1

FIG. 1. Side views of shortenings during compression. A roll-in shortening deforms plastically without forming cracks, whereas an all-purpose shortening displays macroscopic cleavage fracture.

So far, a direct measure of these properties has been impossible, partly because of continued reliance on subjective or empirical tests established over half a century ago [1]. These tests have been used to determine “hardness,” “spreadability," the “consistency” index, and similar attributes typically found in the industry lexicon. The challenge in using such tests lies not so much in their ill-defined physical properties, as in the degree to which such properties can be said to exist at all. This motivated the search for alternative tests or physical measures of roll-in functionality.

Considering that roll-in shortenings (and fats in general) experience substantial deformations during use, it seemed relevant to investigate their nonlinear rheological behavior. For this purpose, large amplitude oscillatory shear (LAOS) tests were conducted and applied to understand yielding dynamics of a wide range of soft matter and complex fluids. Unlike more traditional large deformation tests (cone penetrometry or compression), LAOS oscillations provide controlled yielding, allow the decomposition of elastic and viscous moduli, and better resemble the flow experienced in use as they probed strains and frequencies relevant to lamination. For example, under small amplitude oscillatory shear (SAOS), bakery shortenings behave in a similar fashion; i.e., both fats act as soft viscoelastic solids (G' >G" ) with similar viscoelastic moduli, critical strains, and stresses uncorrelated to solid fat content. Rohm and Weidinger [2] reached similar conclusions in commercial butters. Nevertheless, visualization of the same data in strain versus stress curves makes it possible to realize rheological differences between samples.

Beyond γcritical ≈ 0.05%, both shortenings reached yield stresses (taken as stress maximums) of τ1≈4000Pa (roll-in) and τ1≈ 4900 (all-purpose), respectively. Above the yield stress (failure point), the stress in roll-in exhibits a plateau compared to an all-purpose shortening in which the stress undergoes abrupt decrease and quick stress relaxation. Strong stress overshoots result from the growth of macroscopic fractures in the nonlinear regime, as shown in Fig 1.

To capture the observed behavior, several methods have been proposed to analyze the nonlinear LAOS response. These include Lissajous-Bowditch (L-B) curves and Fourier-transform (FT) rheology, and stress-decomposition via FT-Chebyshev polynomials, among others. For a comprehensive review on the fundamentals, analysis, and applications of the technique, readers can consult the work of Hyun et. al. [3], while readers wanting to know more about use of the technique in lipid materials may refer to Macias-Rodriguez and Marangoni [4].

L-B are parametric plots of strain versus stress (elastic perspective) or strain-rate versus stress (viscous perspective) that appear as ellipsoidal shapes in the linear regime, but progressively distort in the nonlinear regime (Fig. 2). In the nonlinear regime, both shortenings display similar features; i.e., stress “upturns” and stress “bends” within an oscillatory cycle, specially at maximum strains or shear rates (local measures). However, it seems that a roll-in shortening experiences less local strain stiffening; i.e., milder stress upturn, and less shear-thinning, higher viscosity as seen by larger area enclosed by the elastic L-B figures, and weaker stress bending shown in the viscous L-B figures.

Figure 2

FIG 2. (a) Viscoelastic moduli G' and G" as a function of strain input g0. Enlarged data points reflect average responses (G', G") calculated from the raw waveforms. (b) Stress versus strain as obtained from (a) for strain deformations γ0≤15%.  (c) Curves of stress versus strain-elastic projection, and shear rate-viscous projection recorded at w= 3.6 rads-1and g0=0.01-10 % for laminating and all-purpose shortenings. Arrows highlight the observed qualitative differences. Adapted from [5].

Quantification of these responses can be done via a Fourier transform (FT) analysis to reconstruct the LAOS waveforms. In the linear regime, Fourier series show the presence of one or “fundamental” harmonic (n=1), whereas in the nonlinear regime, higher-order odd harmonics (n=1, 3, 5…) grow unboundly in the nonlinear regime. Fig 2 shows the intensity of the leading-order harmonic (n=3) as a function of strain. Compared to an all-purpose shortening, roll-in shortenings gradually transition into the nonlinear regime, suggesting superior ability to bear stresses for the primer.

To gain physical insight into the nonlinear regime, we use the FT-Chebyshev framework to isolate storage- and loss- energy mechanisms in the stress response (Fig. 3). While minor differences appear in the local elastic response (GL'-GM')/ GL', substantial variations, such as dynamic viscosities at minimum- (hM') and maximum- (hL')  shear rates, show in the viscous response (viscosities are related to plastic flow). To illustrate, a roll-in shortening exhibits at least twice the relative dynamic viscosities than an all-purpose shortening does. Increased viscosities mean that a roll-in shortening dissipates energy more effectively during deformation than an all-purpose shortening does, and thus the primer withstands higher stresses than is the case when using all-purpose shortening. Remarkably, it has been found that multiple formulations (e.g. palm-oil based, soybean-oil based, trans-containing, trans free) meet the same rheological criteria. This suggests that novel and functional roll-in shortenings may be designed as long as the rheological properties are adequately matched.

Figure 3

FIG. 3. Nonlinear elastic and viscous local measures at ω= 3.6 rads-1and γ0=0.01-10 % for roll-in and all-purpose shortenings. Viscous measures are parametrized by the linear dynamic viscosity ηLVE' at γ0= 0.01%. Inset shows absolute values of dynamic viscosities. Adapted from [5].

 Braulio A. Macias-Rodriguez is a postdoctoral fellow in the Department of Food Science, University of Guelph (Ontario, Canada). He may be contacted at bmaciasr@uoguelph.ca.
 Alejandro Marangoni is a professor in the Department of Food Science, University of Guelph (Ontario, Canada). He may be contacted at amarango@uoguelph.ca.

Further reading

  1. Haighton, A.J. (1959) The measurement of the hardness of margarine and fats with cone penetrometers. J. Am. Oil Chem. Soc. 36: 345-348. http://dx.doi.org/10.1007/BF02640051
    Rohm, H. and K.H. Weidinger (1993) Rheological behaviour of butter at small deformations. J. Texture Stud. 24: 157-172. http://dx.doi.org/10.1111/j.1745-4603.1993.tb00041.x
  2. Hyun, K, M. Wilhelm, and C.O. Klein, et al. (2011) Progress in polymer science a review of nonlinear oscillatory shear tests : analysis and application of large amplitude oscillatory shear ( LAOS ). Prog. Polym. Sci. 36: 1697-1753. http://dx.doi.org/10.1016/j.progpolymsci.2011.02.002
  3. Macias-Rodriguez, B.A. and A.G. Marangoni (2017) Linear and nonlinear rheological behavior of fat crystal networks. Crit. Rev. Food Sci. Nutr. 8398: 0-0. http://dx.doi.org/10.1080/10408398.2017.1325835
  4. Macias-Rodriguez, B.A, Marangoni AG (2017) Understanding the functionality of lipid-based materials under large-amplitude nonlinear deformations. Lipid Technol. 29: 23-27. http://dx.doi.org/10.1002/lite.201700008