Integral fractional viscoelastic models in SPH: LAOS simulations versus experimental data
Extracted source text · arXiv 2507.14149
%Version 31 May 2025
% See section 11 of the User Manual for version history
%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%% %%
%% Please do not use \input{...} to include other tex files. %%
%% Submit your LaTeX manuscript as one .tex document. %%
%% %%
%% All additional figures and files should be attached %%
%% separately and not embedded in the \TeX\ document itself. %%
%% %%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%\documentclass[referee,sn-basic]{sn-jnl}% referee option is meant for double line spacing
%%=======================================================%%
%% to print line numbers in the margin use lineno option %%
%%=======================================================%%
%%\documentclass[lineno,sn-basic]{sn-jnl}% Basic Springer Nature Reference Style/Chemistry Reference Style
%%======================================================%%
%% to compile with pdflatex/xelatex use pdflatex option %%
%%======================================================%%
%%\documentclass[pdflatex,sn-basic]{sn-jnl}% Basic Springer Nature Reference Style/Chemistry Reference Style
%%Note: the following reference styles support Namedate and Numbered referencing. By default the style follows the most common style. To switch between the options you can add or remove Numbered in the optional parenthesis.
%%The option is available for: sn-basic.bst, sn-vancouver.bst, sn-chicago.bst%
%%\documentclass[pdflatex,sn-nature]{sn-jnl}% Style for submissions to Nature Portfolio journals
%%\documentclass[pdflatex,sn-basic]{sn-jnl}% Basic Springer Nature Reference Style/Chemistry Reference Style
\documentclass[pdflatex,sn-mathphys-num]{sn-jnl}% Math and Physical Sciences Numbered Reference Style
%%\documentclass[pdflatex,sn-mathphys-ay]{sn-jnl}% Math and Physical Sciences Author Year Reference Style
%%\documentclass[pdflatex,sn-aps]{sn-jnl}% American Physical Society (APS) Reference Style
%%\documentclass[pdflatex,sn-vancouver,Numbered]{sn-jnl}% Vancouver Reference Style
%%\documentclass[pdflatex,sn-apa]{sn-jnl}% APA Reference Style
%%\documentclass[pdflatex,sn-chicago]{sn-jnl}% Chicago-based Humanities Reference Style
%%%% Standard Packages
%%<additional latex packages if required can be included here>
\usepackage{graphicx}%
\graphicspath{{images/}} % Add this line to specify the folder path for images
\usepackage{multirow}%
\usepackage{amsmath,amssymb,amsfonts}%
\usepackage{amsthm}%
\usepackage{mathrsfs}%
\usepackage[title]{appendix}%
\usepackage{xcolor}%
\usepackage{textcomp}%
\usepackage{manyfoot}%
\usepackage{booktabs}%
\usepackage{algorithm}%
\usepackage{algorithmicx}%
\usepackage{algpseudocode}%
\usepackage{listings}%
%%%%
\usepackage{subcaption}
\usepackage{gensymb} % Add this line to use the \degree C symbol
% COLORE PAGINA: COMMENTAMI SE NON TI PIACE
% \pagecolor[rgb]{0.1, 0.1, 0.1}
% \color[rgb]{0.6, 0.6, 0.6}
%%%%%=============================================================================%%%%
%%%% Remarks: This template is provided to aid authors with the preparation
%%%% of original research articles intended for submission to journals published
%%%% by Springer Nature. The guidance has been prepared in partnership with
%%%% production teams to conform to Springer Nature technical requirements.
%%%% Editorial and presentation requirements differ among journal portfolios and
%%%% research disciplines. You may find sections in this template are irrelevant
%%%% to your work and are empowered to omit any such section if allowed by the
%%%% journal you intend to submit to. The submission guidelines and policies
%%%% of the journal take precedence. A detailed User Manual is available in the
%%%% template package for technical guidance.
%%%%%=============================================================================%%%%
%% as per the requirement new theorem styles can be included as shown below
\theoremstyle{thmstyleone}%
\newtheorem{theorem}{Theorem}% meant for continuous numbers
%%\newtheorem{theorem}{Theorem}[section]% meant for sectionwise numbers
%% optional argument [theorem] produces theorem numbering sequence instead of independent numbers for Proposition
\newtheorem{proposition}[theorem]{Proposition}%
%%\newtheorem{proposition}{Proposition}% to get separate numbers for theorem and proposition etc.
\newcommand{\gammadot}{\dot{\gamma}}
\raggedbottom
%%\unnumbered% uncomment this for unnumbered level heads
%%%%% PERSONALISATION %%%%%
\usepackage{physics}
\usepackage[normalem]{ulem}
\newcommand{\vc}[1]{\boldsymbol{#1}}
\newcommand{\vt}[1]{\boldsymbol{\mathsf{#1}}}
\newcommand{\vv}[1]{\mathsf{\boldsymbol{#1}}}
\newcommand\bluecirc{\ensuremath{{\color{blue}\bullet}}}
\newcommand\bluediam{\ensuremath{{\color{blue}\blacklozenge}}}
% Custom command for comments by Luca
\newcommand{\luca}[1]{\textcolor{red}{\emph{\underline{Luca}}: \textsf{#1}}}
\begin{document}
\title[Integral-fractional SPH LAOS]{Integral fractional viscoelastic models in SPH: LAOS simulations versus experimental data}
%%=============================================================%%
%% GivenName -> \fnm{Joergen W.}
%% Particle -> \spfx{van der} -> surname prefix
%% FamilyName -> \sur{Ploeg}
%% Suffix -> \sfx{IV}
%% \author*[1,2]{\fnm{Joergen W.} \spfx{van der} \sur{Ploeg}
%% \sfx{IV}}\email{iauthor@gmail.com}
%%=============================================================%%
\author*[1]{\fnm{Luca} \sur{Santelli}}\email{lsantelli@bcamath.org}
\author[2]{\fnm{Adolfo} \sur{Vázquez Quesada}}\email{a.vazquez-quesada@fisfun.uned.es}
% \equalcont{These authors contributed equally to this work.}
\author[3]{\fnm{Aizeti} \sur{Burgoa}}\email{aburgoa@leartiker.com}
\author[3]{\fnm{Aitor} \sur{Arriaga}}\email{aburgoa@leartiker.com}
\author[3]{\fnm{Rikardo} \sur{Hernandez}}\email{rhernandez@leartiker.com}
\author[1,4,5]{\fnm{Marco} \sur{Ellero}}\email{mellero@bcamath.org}
% \equalcont{These authors contributed equally to this work.}
\affil*[1]{\orgdiv{} \orgname{Basque Center for Applied Mathematics}, \orgaddress{\street{Mazarredo Zumarkalea, 14}, \city{Bilbao}, \postcode{48009}, \state{Bizkaia}, \country{Spain}}}
\affil[2]{\orgdiv{Facultad de Ciencias}, \orgname{Universidad Nacional de Educación a Distancia}, \orgaddress{\street{ C. de Bravo Murillo, 38}, \city{Madrid}, \postcode{28015}, \country{Spain}}}
\affil[3]{Sustainable Transport group, Polymers Technology, Leartiker S. Coop, \city{Markina-Xemein}, \postcode{48270}, \country{Bizkaia (Spain)}}
\affil[4]{\orgdiv{IKERBASQUE}, \orgname{Basque Foundation for Science}, \orgaddress{\street{Calle de María Díaz de Haro 3}, \city{Bilbao}, \postcode={48013}, \country{Spain}}}
\affil[5]{\orgdiv{Complex Fluid Group, Department of Chemical Engineering}, \orgname{Swansea University}, \orgaddress{\street{SA1 8EN}, \city{Swansea}, \country{United Kingdom}}}
%%==================================%%
%% Sample for unstructured abstract %%
%%==================================%%
\abstract{
The rheological behaviour of a polymer was investigated by performing numerical simulations in complex flow and comparing them to experiments. For our simulations, we employed a Smoothed Particle Hydrodynamics scheme, utilizing an integral fractional model based on the K-BKZ framework. The results are compared with experiments performed on melt-state isotactic polypropylene under medium and large amplitude oscillatory shear.
The numerical results are in good agreement with the experimental data, and the model is able to capture and predict both the linear and the non-linear viscoelastic behaviours of the polymer melt. Results show that equipping SPH with an integral fractional model is promising approach for the simulation of complex polymeric materials under realistic conditions.
}
%%================================%%
%% Sample for structured abstract %%
%%================================%%
% \abstract{\textbf{Purpose:} The abstract serves both as a general introduction to the topic and as a brief, non-technical summary of the main results and their implications. The abstract must not include subheadings (unless expressly permitted in the journal's Instructions to Authors), equations or citations. As a guide the abstract should not exceed 200 words. Most journals do not set a hard limit however authors are advised to check the author instructions for the journal they are submitting to.
%
% \textbf{Methods:} The abstract serves both as a general introduction to the topic and as a brief, non-technical summary of the main results and their implications. The abstract must not include subheadings (unless expressly permitted in the journal's Instructions to Authors), equations or citations. As a guide the abstract should not exceed 200 words. Most journals do not set a hard limit however authors are advised to check the author instructions for the journal they are submitting to.
%
% \textbf{Results:} The abstract serves both as a general introduction to the topic and as a brief, non-technical summary of the main results and their implications. The abstract must not include subheadings (unless expressly permitted in the journal's Instructions to Authors), equations or citations. As a guide the abstract should not exceed 200 words. Most journals do not set a hard limit however authors are advised to check the author instructions for the journal they are submitting to.
%
% \textbf{Conclusion:} The abstract serves both as a general introduction to the topic and as a brief, non-technical summary of the main results and their implications. The abstract must not include subheadings (unless expressly permitted in the journal's Instructions to Authors), equations or citations. As a guide the abstract should not exceed 200 words. Most journals do not set a hard limit however authors are advised to check the author instructions for the journal they are submitting to.}
\keywords{Smoothed Particle Hydrodynamics, Large Amplitude Oscillatory Shear, Polymer Melts, Integral Fractional Model}
%%\pacs[JEL Classification]{D8, H51}
%%\pacs[MSC Classification]{35A01, 65L10, 65L12, 65L20, 65L70}
\maketitle
\section{Introduction}\label{sec:introduction}
Large Amplitude Oscillatory Shear (LAOS) is a powerful rheological technique used to investigate the nonlinear viscoelastic properties of complex fluids, such as polymer melts, colloidal suspensions, and biological materials. Unlike Small Amplitude Oscillatory Shear (SAOS), which probes the linear viscoelastic regime, LAOS applies larger deformations to the material, allowing researchers to explore the material's response under conditions that are more representative of real-world applications \cite{hyun2011review,yosick1997kinetic}. The technique provides detailed insights into the material's structure and dynamics by analyzing the stress response as a function of strain amplitude and frequency. This information is crucial for understanding the material's behaviour under more realistic processing and end-use conditions, enabling the development of better-performing materials and products. LAOS experiments can reveal phenomena such as strain hardening, strain softening, thixotropy and yielding, which are not accessible through linear viscoelastic measurements \cite{blackwell2014simple,koekuti2016exploring,vargas2023multiscale}. Consequently, LAOS has become an essential tool in the field of rheology for characterizing the complex mechanical behavior of advanced materials.
A proper description of LAOS requires a rheological model that is able to capture the intrinsic non-linear behaviour of the complex fluid when subjected to large deformations \cite{hyun2011review,vargas2023multiscale}. A number of these descriptions relies on the framework provided by the integral Kaye–Bernstein, Kearsley, Zapas (K-BKZ) model \cite{bird1987dynamics,owens2002computational}. The model is based on the idea that the stress response of a material can be expressed as a convolution of the material's relaxation modulus with the strain history. Indeed, it has been shown that a polymer can be naturally described by employong a generalised integral-fractional K-BKZ constitutive model that captures both the power-law behaviour, typical of polymeric materials, and the non-linear response induced by the flow \cite{jaishankar2014fractional,santelli2024smoothed}. Fractional models describe the rheological behaviour of a fluid by introducing a constitutive element known as the spring-pot, which interpolates the elastic response of a spring and the viscous response of a dashpot \cite{gemant1938fractional,blair1942classification,jaishankar2013power,jaishankar2014fractional,yang2010constitutive,santelli2024smoothed}. The matematical rapresentation of the spring-pot is given by the fractional derivative, which is a generalization of the classical derivative that allows for non-integer order derivatives.
However, solving the K-BKZ equation analytically (with or without the fractional formulation) can be challenging, due to the non-linearities and the complex conditions that arise in LAOS experiments. As a result, numerical simulations have become an essential tool for studying the rheological behaviour of complex fluids under LAOS.
Notably, the fractional derivative can be solved by an integral equation, naturally suggesting the adoption of an integral Lagrangian model for the numerical scheme \cite{larson1999structure,santelli2024smoothed}.
While integral models present several advantages, such as the ability to capture the long-term memory effects of complex fluids, they also come with challenges. Indeed, the integral equation can be computationally expensive and time-consuming, especially due to the need of tracking particle position \cite{viriyayuthakorn1980finite,rasmussen1993simulation,keunings2003finite,owens2002computational}.
A possible way to approach this issue is to perform simulations using a numerical method written in a Lagrangian formulation, thus overcoming the particle history tracking problem. We have shown that the Smoothed Particle Hydrodynamics method (SPH) is well equipped for this task, as it is a Lagrangian particle-based method that can handle large deformations and complex geometries, including constitutive equations written in integral fractional form \cite{ellero2002viscoelastic,ellero2005sph,vazquezquesada2009smoothed, santelli2024smoothed}, while experiencing only a minor loss in performance and accuracy.
%The rheological behavior of complex fluids under LAOS is typically described by a constitutive model that captures the material's viscoelastic properties. One of the most widely used models for describing the nonlinear viscoelastic behavior of polymer melts is the K-BKZ model. This model is based on the Oldroyd-B framework and includes a series of nonlinear differential equations that describe the stress response of the material under large deformations. The K-BKZ model has been successfully applied to a wide range of complex fluids, including polymer melts, colloidal suspensions, and biological materials. However, solving the K-BKZ equations analytically can be challenging due to the nonlinear nature of the equations and the complex boundary conditions that arise in LAOS experiments. As a result, numerical simulations have become an essential tool for studying the rheological behavior of complex fluids under LAOS.
In this article, we investigate the rheological behavior of a polymer melt under LAOS using numerical simulations based on the Smoothed Particle Hydrodynamics method. This manuscript is organized as follows:
in Section 2, we discuss the theoretical framework behind the integral approach and LAOS rheology; in Section 3, we describe the experimental setup and the numerical model used in this study; in Section 4, we present the results of the simulations and compare them to experimental data; and in Section 5, we discuss the implications of our findings and suggest future research directions.
\section{Theoretical integral framework}\label{sec:theory}
%\emph{Notation}.
The integral rapresentation of the constitutive equation can be understood as a simple mathematical transformation over the more common differential approach; however, its mathematical formulation is not-trivial. We report here our implementation, which strictly mimic the construction of \cite{bird1987dynamics}.
The main variable of this analysis is the displacement function $\vb{r} = \vb{r}(\vb r', t', t)$, which tracks the current position $\boldsymbol{r}$ of a particle, at time $t$, as a function of its past position $\vb r'$ at time $t'$. In this representation, the pair $\vb r', t'$ identifies the particle we are tracking. The Cartesian components of $\boldsymbol{r}$ are identified by $x_\mu$, with the Greek index ranging between the number of dimensions.
The velocity field can be directly derived by the displacements:
$\vb v'(\vb r',t')= \pdv{\vb r'(\vb r,t,t')}{t'}$ and $\vb v(\vb r,t)= \pdv{\vb r(\vb r',t',t)}{t}$.
From the displacement function we define a displacement gradient
\begin{equation}
\Delta_{\mu\nu} (\vb r , t, t') = \pdv{x'_\mu(\vb r,t,t')}{x_\nu}
\label{eq:grad_r'_bird}
\end{equation}
i.e. the gradient of the past position of the fluid particles at time $t'$ around the current position $\vb r$.
The equivalent form in terms of displacement of the current position around the previous is
\begin{equation}
E_{\mu\nu} (\vb r , t, t') = \pdv{x_\mu(\vb r',t',t)}{x'_\nu}.
\end{equation}
Together, they satisfy $\sum_{\mu,\nu}\Delta_{\mu\nu}E_{\mu\nu}=\delta_{\mu\nu}$ and, for incompressible fluids, $\det(\Delta_{\mu\nu})=\det(E_{\mu\nu})=1$.
From $\vb \Delta$ and $\vb E$, we construct, respectively, the Cauchy ($\vt B^{-1} $) and Finger ($\vt B$) finite strain tensors, the components of which are defined as follows:
\begin{align}
&B^{-1}_{\mu\nu}(\vb r,t,t') = \sum_\xi\pdv{x'_\xi}{x_\mu}\pdv{x'_\xi}{x_\nu}\\
&B_{\mu\nu}(\vb r,t,t') = \sum_\xi\pdv{x_\xi}{x'_\mu}\pdv{x_\xi}{x'_\nu}.
\end{align}
These are symmetric positive definite tensors, which are invariant under rigid rotation and translation. Thus they can be used to build finite strain tensors that vanish for rigid body motions:
\begin{align}
&\boldsymbol{\gamma}^{[0]} = \vt B^{-1}-\vt 1\\
&\boldsymbol{\gamma}_{[0]} = \vt 1 - \vt B
\end{align}
where $\vt 1$ is the identity tensor and $\boldsymbol{\gamma}^{[0]}$ and $\boldsymbol{\gamma}_{[0]}$ are the so-called relative finite strain tensors.
We note that, while for small displacements $\boldsymbol{\gamma}^{[0]}=\boldsymbol{\gamma}_{[0]}$, this is, in general, not true for large displacements. Comparison with experimental data shows that more realistic constitutive models, e.g. the corresponding integral version of the upper convected Maxwell, are built upon the finite strain tensor $\boldsymbol{\gamma}_{[0]}$.
The principal invariants of the Finger strain tensor $\vt B$, defined as the coefficients of its characteristic polynomial, are
\begin{equation}
I_1=\tr\vt B=\tr(\vt 1-\vt\gamma_{[0]}),
\end{equation}
\begin{equation}
I_2=\tfrac12[(\tr\vt B)^2-\tr\vt B^2]=\tr\vt B^{-1}=\tr(\vt 1+\vt\gamma^{[0]}),
\end{equation}
and $I_3=\det\vt B$, which, for an incompressible fluid, equals one.
These scalars are used to construct a frame-indifferent model.
\subsection{Fractional K--BKZ model}
For the constitutive prescription of the incompressible viscoelastic fluid, we adopt the formulation of the fractional K--BKZ model. The classical K--BKZ model assumes the existence of a function $V$, scalar in its arguments $t-t',I_1,I_2$, and takes the nonlinear single-integral equation for the stress tensor of the form
\begin{equation}
\vt \tau(t) = \int_{-\infty}^t
\left[
\frac{\partial V(t-t',I_1,I_2)}{\partial I_1}\vt\gamma_{[0]}(t,t')+
\frac{\partial V(t-t',I_1,I_2)}{\partial I_2}\vt\gamma^{[0]}(t,t')
\right]
\dd t'.
\label{eq:K--BKZ}
\end{equation}
It is customarily assumed separability of $V$ in time-dependent and strain-dependent factors:
\begin{equation}
V(t-t',I_1,I_2) = M(t-t')W(I_1,I_2)
\label{eq:}
\end{equation}
with $M$ the so-called \emph{memory} function and $W$ the elastic potential function, with the consistency condition with the linear limit $\left(\frac{\partial W}{\partial I_1}\right)_{3,3}+\left(\frac{\partial W}{\partial I_2}\right)_{3,3}=1$. To guarantee well-posedness for finite positive $M$, it is also required that $W$ is monotone in each argument, strictly monotone in $I_1$ or $I_2$, and convex in $\sqrt{I_1}$ and $\sqrt{I_2}$. The memory function $M$ is related to the so-called \emph{relaxation modulus} $G$ via the relation
\begin{equation}
M(t-t')=\frac{\partial G(t-t')}{\partial t'}.
\end{equation}
The factorized K--BKZ equation reads
\begin{equation}
\vt \tau(t) = \int_{-\infty}^tM(t-t')
\left[
\frac{\partial W(I_1,I_2)}{\partial I_1}\vt\gamma_{[0]}(t,t')+
\frac{\partial W(I_1,I_2)}{\partial I_2}\vt\gamma^{[0]}(t,t')
\right]
\dd t'.
\label{eq:K--BKZ_factorized}
\end{equation}
In the following, we neglect the dependence of $W$ on $I_2$, so that the second addend in the integral vanishes. This is constistent with the previous claim of choosing $\gamma_{[0]}$ as the preferred option for the finite strain tensor.
\subsubsection{Shear deformation}
We consider the unsteady homogeneous simple shear flow for a fluid confined between two parallel planes. We write $\tau$ for the shear component $\tau_{yx}$ of the stress tensor and $\gamma$ for the shear component $\gamma_{yx}$ of the strain tensor $\vt{\gamma}_{[0]}$. In this regime, the components of the velocity of the fluid are then $v_x=\dot{\gamma}(t)y$, $v_y=0$, $v_z=0$. We introduce a function $h$, called \emph{damping} function, defined by $h(\gamma)=\tfrac{\partial W(I_1)}{\partial I_1}$, where we used the fact that, for this flow, $I_1=3+\gamma^2$.
The equation for the shear component of the stress can then be written as
\begin{equation}
\vt \tau(t) = \int_{-\infty}^tM(t-t')
h(\gamma)\vt\gamma(t,t')
\dd t'
\end{equation}
where
\[
\gamma(t,t')=\int_t^{t'}\dot{\gamma}(t'')\dd t''=\gamma(t')-\gamma(t).
\]
%which, for a steady flow, simplifies in $\gamma(t,t')=\dot{\gamma}(t'-t)$.
We take
\begin{equation}
h(\gamma)=\frac{1}{1+a\gamma^b}
\label{eq:damping_function}
\end{equation}
with $a,b$ some fitting parameters, which need to be determined experimentally.
We adopt the Fractional Maxwell Model for the shear stress-strain relation, which consists of two spring-pots, each characterized by a pair of material parameters, denoted by $\mathbb V$, $\alpha$ and $\mathbb G$, $\beta$, respectively, arranged in series \cite{jaishankar2014fractional}. Namely,
\begin{equation}
\tau(t) + \frac{\mathbb V}{\mathbb G}\dv[\alpha-\beta]{\tau(t)}{t}=\mathbb V \dv[\alpha]{\gamma(t)}{t}
\end{equation}
where the Caputo fractional derivative of order $\beta$ in time is defined as
\begin{equation}
\dv[\beta]{\vt\gamma(t)}{t} \equiv \frac{1}{\Gamma(1-\beta)}\int_0^t (t-t')^{-\beta}\dot{\vt\gamma}(t')\dd t'
\end{equation}
with $\Gamma(\cdot)$ the complete Gamma function and $0\le\beta\le 1$.
The relaxation modulus and memory function associated with the Fractional Maxwell Model are then
\begin{align}
&G(t-t') = \mathbb G (t-t')^{-\beta}E_{\alpha-\beta, 1-\beta}\qty(-\frac{\mathbb G}{\mathbb V}(t-t')^{\alpha-\beta})\\
&M(t-t') = -\mathbb G (t-t')^{-1-\beta}E_{\alpha-\beta, -\beta}\qty(-\frac{\mathbb G}{\mathbb V}(t-t')^{\alpha-\beta})
\label{eq:memory_fractional}
\end{align}
where $E_{a,b}$ is the generalized Mittag-Leffler function, given by
\begin{equation}
E_{a,b}(z) = \sum_{k=0}^\infty \frac{z^k}{\Gamma(ak+b)}.
\end{equation}
In particular, $G(t)\to0$ for $t\to+\infty$ and $M(t_1)<M(t_2)$ for $t_1<t_2$.
Ultimately, our simulations ground on the following constitutive equation:
\begin{equation}
\vt \tau(t) = \int_{-\infty}^t
\!\left[-\mathbb G (t-t')^{-1-\beta}E_{\alpha-\beta, -\beta}\qty(-\frac{\mathbb G}{\mathbb V}(t-t')^{\alpha-\beta})\right]
\frac{1}{1+a\gamma(t')^b}(\gamma(t')-\gamma(t))
\dd t'.
\label{eq:K-BKZ_fractional}
\end{equation}
\subsubsection{Oscillatory shear}
We now consider oscillatory shearing strain input of the form $\gamma(t)=\gamma_0\sin(\omega t)$, with corresponding shear rate $\dot{\gamma}(t)=\omega\gamma_0\cos(\omega t)$. Experiments testing such a flow are of two types, which differ for the magnitude of the amplitude $\gamma_0$ of the sinusoidal oscillations: the \emph{small amplitude oscillatory shear (SAOS)} and the \emph{large amplitude oscillatory shear (LAOS)} experiments. The former is tipically conducted at a fixed small amplitude (approximately $\gamma_0\sim10^{-2}$) and measures the amplitude and phase shift of the shear stress as a function of the frequency $\omega$ and allows to characterise the nearly linear behaviour of the fluid. In contrast, in the latter, the oscillations are applied at a fixed frequency with increasing amplitudes and allow quantifying the viscoelastic nonlinear response. Typical values for the strain amplitudes required to go beyond the linear regime are of the order $\gamma_0\sim 10^0$.
%...
In the nonlinear regime, the storage modulus $G'(\omega)$ and the loss modulus $G''(\omega)$ aquire an additional dependency on the input strain amplitude $\gamma_0$ and lose their precise physical meaning of elastic and viscous components. This additional dependency can be easily understood once we consider that these moduli are built from the Fourier transformation of the shear stress response to a sinusoidal strain input, and when large deformations are applied the response is not purely sinusoidal, but rather a superposition of harmonics of higher frequencies.
Therefore, the shear stress output can be represented as a linear superposition of harmonic contributions and be taken of the form \cite{hyun2011review}
\begin{equation}
\vt\tau(t)=\sum_{p,odd}\sum_{q,odd}^p
\gamma_0^q\left(a_{pq}\sin(q\omega t)+b_{pq}\cos(q\omega t)\right)
\end{equation}
which is often referred to as a power series \cite{giacomin2011large}. It is important to note that the series contains only the odd harmonics of the shear deformations, and $a_{11}=G'(\omega)$, $b_{11}=G''(\omega)$ are the linear moduli.
An alternative way to represent $\vt \tau(t)$, named Fourier series, reads
\begin{equation}
\label{eq:fourier_series}
\vt\tau(t)=\gamma_0\sum_{n,odd}\left(G'_{n}(\omega, \gamma_0)\sin(n\omega t)+G''_{n}(\omega, \gamma_0)\cos(n\omega t)\right)
\end{equation}
and this will be the preferred form in the rest of the manuscript. In this representation, the coefficients $G'_{n}(\omega, \gamma_0)$ and $G''_{n}(\omega, \gamma_0)$ are the Fourier coefficients of the stress response, which depend on both the frequency $\omega$ and the strain amplitude $\gamma_0$. The first harmonic corresponds to the linear moduli: $G'(\omega)=G'_{1}(\omega, \gamma_0)$ and $G''(\omega)=G''_{1}(\omega, \gamma_0)$.
\emph{SAOS limit}.
The linear viscoelastic response of a material described by the fractional K-BKZ model can be obtained by the following procedure.
If we analyse a system described by a single spring-pot and apply a step strain $\gamma(t) = \gamma_0 H(t)$, with $\gamma_0$ sufficiently small, it results in a power-law decay of the stress:
\begin{equation}
\tau(t)/\gamma_0 \equiv G(t) = \frac{\mathbb{G}}{\Gamma(1-\beta)}t^{-\beta}
\label{eq:relax_modulus_singleSP_time}
\end{equation}
which is the response of a class of material known as critical gel \cite{rathinaraj2021incorporating}. The behaviour under SAOS can be computed by Fourier transformation of its constitutive Equation, obtaining
\begin{align}
&G'(\omega) = \mathbb{G}\cos(\pi\beta/2)\omega^\beta\\
&G''(\omega) = \mathbb{G}\sin(\pi\beta/2)\omega^\beta.
\end{align}
The response of the Fractional Maxwell Model is obtained by the same procedure and results in the following expressions for the linear moduli:
\begin{align}
&G'(\omega) = \frac{ (\mathbb{G}\omega^\beta)^2{\mathbb V\omega^\alpha}\cos{(\pi\alpha/2)}+(\mathbb{V}\omega^\alpha)^2 \mathbb G\omega^\beta\cos{(\pi\beta/2)}}
{(\mathbb{V}\omega^\alpha)^2+(\mathbb{G}\omega^\beta)^2+2\mathbb V \mathbb G \omega^{\alpha+\beta}\cos{[\pi(\alpha-\beta)/2]}}\\
&G''(\omega) = \frac{ (\mathbb{G}\omega^\beta)^2{\mathbb V\omega^\alpha}\sin{(\pi\alpha/2)}+(\mathbb{V}\omega^\alpha)^2 \mathbb G\omega^\beta\sin{(\pi\beta/2)}}
{(\mathbb{V}\omega^\alpha)^2+(\mathbb{G}\omega^\beta)^2+2\mathbb V \mathbb G \omega^{\alpha+\beta}\cos{[\pi(\alpha-\beta)/2]}}
\end{align}
\section{Experimental and numerical setups}\label{sec:experimental}
\subsection{Experimental setup}\label{sec:experimental_setup}
\subsubsection{Materials}
The material selected for the present study was an unreinforced isotactic Polypropylene (iPP) homopolymer, supplied by Borealis (Linz, Austria). The specific grade used was HF955MO, which is a widely used thermoplastic polymer in various industrial applications. Table \ref{tab:material_properties} summarizes the basic mechanical and physical properties of the material, as provided by the resin manufacturer.
Additionally, the material supplier specified an average molecular weight (Mw) of $206$ kDalton.
\begin{table}[htbp]
\centering
\begin{tabular}{|c|c|c|}
\hline
\textbf{Property} & \textbf{Typical value} & \textbf{Test method} \\
\hline
Density & 0.905 Tn/mm\textsuperscript{3} & ISO 1183 \\
\hline
Melt flow rate (230 °C/2.16 Kg) & 20 g/10 min & ISO 1133-1 \\
\hline
Tensile modulus (1 mm/min) & 2200 MPa & ISO 527-2 \\
\hline
Tensile strain at yield (50 mm/min) & 6 \% & ISO 527-2 \\
\hline
Tensile stress at yield (50 mm/min) & 40 MPa & ISO 527-2 \\
\hline
Heat deflection temperature B (0.45 MPa) & 115 ºC & ISO 75-2 \\
\hline
\end{tabular}
\caption{Summary of the physical and thermo-mechanical properties of the selected isotactic Polypropylene (iPP).}
\label{tab:material_properties}
\end{table}
\begin{figure}[b!]
\centering
\includegraphics[width=0.8\textwidth]{tandelta_vs_temperature.png}
\caption{Identification of the $\tan{\delta}$ peak for the iPP material to determine the glass transition temperature (Tg).}
\label{fig:tandelta_vs_temperature}
\end{figure}
\subsubsection{Experimental methods}
Prior to the melt-state rheological tests, which were the target experiments for further constructing the modelling approach, a series of fundamental measurements were carried out in solid state, i.e. Dynamic Mechanical Thermal Analysis (DMTA) was used to identify the glass transition temperature (Tg) of the iPP, while Differential Scanning Calorimetry (DSC) provided insights into the crystallinity of the injection moulded granules of the material and facilitated the determination of the melting temperature (Tm).
\paragraph{Dynamic Mechanical Thermal Analysis (DMTA) }
A Metravib $+300$ instrument from Acoem (Limonest, France) was used in tensile mode to identify the glass transition temperature (Tg) of the selected Polypropylene. Rectangular coupons, measuring $20$ mm in height and $6$ mm in width, were cut from injection-moulded plaques with a thickness of $2$ mm. The temperature range for the Tg identification was set between $-40$ and $+100$\degree C, with a heating rate of $5$\degree C/min. A frequency of $10$ Hz was applied, in combination with a dynamic displacement amplitude of $5$ microns.
Figure \ref{fig:tandelta_vs_temperature} shows the plot of the tangent delta $\tan\delta$, which represents the ratio of the loss modulus to the storage modulus of the material, as a function of temperature. A peak in the $\tan\delta$ values around $10$\degree C corresponds to the Tg of the selected isotactic polypropylene homopolymer.
\begin{figure}[tbp]
\centering
\begin{subfigure}{0.45\textwidth}
\centering
\includegraphics[width=\textwidth]{pelletA.jpg}
\caption{~}
\label{fig:pellet1}
\end{subfigure}
\hfill
\begin{subfigure}{0.45\textwidth}
\centering
\includegraphics[width=\textwidth]{pelletB.jpg}
\caption{~}
\label{fig:pellet2}
\end{subfigure}
\vskip\baselineskip
\begin{subfigure}{0.45\textwidth}
\centering
\includegraphics[width=\textwidth]{pelletC.jpg}
\caption{~}
\label{fig:pellet3}
\end{subfigure}
\hfill
\begin{subfigure}{0.45\textwidth}
\centering
\includegraphics[width=\textwidth]{pelletD.jpg}
\caption{~}
\label{fig:pellet4}
\end{subfigure}
\caption{Material in granular state (a,b) and molten state (c,d) prior to testing in oscillatory mode.}
\label{fig:pellets}
\end{figure}
\paragraph{Differential Scanning Calorimetry (DSC)}
A test instrument from TA Instruments (New Castle, USA), model Q 100, was used to determine the melting temperature of the material pellets as well as the crystallinity degree upon cooling. A heating rate of $1\degree C$/min was applied, which resulted in a melting temperature (Tm) of $170\degree C$ and a crystallinity degree of $64\%$. The crystallinity degree was determined using the enthalpy of fusion method, where the heat of fusion of the sample was compared to that of a perfectly crystalline material \cite{lanyi2019determination}. The melting temperature (Tm) was identified as the peak of the endothermic transition in the DSC curve, as shown in figure \ref{fig:DSC}.
\begin{figure}[tbp]
\centering
\includegraphics[width=0.7\textwidth]{heatflow_single.png}
\caption{Heat flow versus temperature, identifying $Tm$ at the peak at $171$°C}
\label{fig:DSC}
\end{figure}
\subsubsection{Melt-state rheology}
The rheological tests performed in the melt state provided the foundational data required for the subsequent modelling of the isotactic polypropylene's flow behaviour. A Hakee Mars III torsional rheometer from Thermo Fisher Scientific (Waltham, USA) was used in order to evaluate both the Small Amplitude Oscillatory Shear (SAOS) response and the Large Amplitude Oscillatory Shear (LAOS) regime of the material.
%Testing temperature was set to $190\degree C$, above the melting temperature identified by DSC, with material pellets placed on the test specimen holder, which had a diameter of $20$ mm. A gap of $0.5$ mm was employed.
Testing temperature was set to 190°C, above the melting temperature identified by DSC, with material pellets placed on the test specimen lower plate, which had a diameter
of $20$ mm. A gap of $0.5$ mm was employed and all the excess material was trimmed before every measurement. Pictures of the material before and after melting are shown in figure \ref{fig:pellets}.
The rheometer test campaign for the LAOS measurements consisted of performing strain amplitude sweeps from $20$ to $500\%$ strain, or $0.01$ to $0.25$ radians at a frequency of $1$Hz ($6.2832$ rad/s). For the SAOS measurements the strain was fixed at $0.1\%$ ($5.10^{-5}$ rad) and the frequencies were varied from $0.1$ to $100$Hz, in order to capture fundamental viscoelastic properties, including shear stiffness (G', G'', G*), compliance (J', J'', $\text{J}^*$), and viscosity ($\eta$', $\eta$'', $\eta^*$). Three different samples were tested for each type of test to ensure a consistent mean response of the material.
%The rheometer test campaign consisted of performing both frequency and strain amplitude sweeps in order to capture fundamental viscoelastic properties, including shear stiffness (G', G'', G*), compliance (J', J'', $\text{J}^*$), and viscosity ($\eta$', $\eta$'', $\eta^*$) %%%%and their evolution with respect to test rate and amplitude.
% Frequency sweeps were conducted from 0.1 to 100 Hz at a fixed strain amplitude of 0.1%.
% Strain amplitude sweeps were performed from 0.1 to 100% at a fixed oscillation frequency of 1 Hz.
%Three different samples were tested for each type of test to ensure a consistent mean response of the material.
\subsection{Numerical implementation}
\label{sec:numerical_implementation}
Simulations of the experiments are implemented using the Smoothed Particle Hydrodynamics (SPH) method, with the structure described in \cite{ellero2010sph, vazquezquesada2009smoothed} and the integral implementation discussed in \cite{santelli2024smoothed}. The main concepts are summarised here.
SPH is a mesh-free Lagrangian model discretising the prescribed equations using a set of $N_p$ fluid particles. Each particle represents a discrete fluid element (labeled by $i,j=1,\dots,N_p$). and is associated to its position and momentum.
%
Mass conservation ensured by defining a particle number density
\begin{equation}
d_i\equiv\sum_j W(|\vb r_{ij}|, r_{cut})
\end{equation}
where (here and thereafter in this section) the sum is intended over all the neighbouring particles of particle $i$. A vector variable with two subscripts is computed as the difference between the values identified by the first and second subscript (e.g. $\vb r_{ij} \equiv \vb r_i - \vb r_j$ is the distance between position $\vb r_i$ and $\vb r_j$).
$W(|\vb r_{ij}|, r_{cut})$ is an interpolating kernel. The choice of the kernel for this work is a quintic spline kernel function with compact support $r_{cut}$, as \cite{morris1997modeling}. Finally, the normalised distance value is defined as $\vb e_{ij}=\vb r_{ij}/ r_{ij}$.
Each particle follows the equation of motion (see \cite{espanol2003smoothed})
\begin{align}
\vb{\dot{r}}_i&= \vb v_i\\
%
m\dot{\vb v_i} &= -\sum_j \qty[\frac{\vc \pi_i}{d_i^2}+ \frac{\vv{\pi}_j }{d_j^2}]W'_{ij}\vb e_{ij}+2\eta_s\sum_j\frac{1}{d_id_j}\frac{W'_{ij}}{r_{ij}} \vb v_{ij}
\label{eq:SPH_vel}
\end{align}
Here, $\eta_s$ is the viscosity of the Newtonian solvent
and $W'_{ij}\equiv\pdv{W(\vb r,r_{cut})}{\vb r} \eval_{\vb r = \vb r_{ij}}$ is the magnitude of the gradient of the smoothing kernel.
$\vc \pi_i$ is the total stress tensor, which can be separated into the extra-stress $\vt \tau_i$ and the isotropic particle pressure tensor $P_i\vt{1}$, where $P_i = p_0[(\rho_i/\rho_0)^\gamma-1]$ and $\rho_0$ and $p_0$ are reference fluid density and pressure respectively while $\gamma=7$ and $c_s=\sqrt{\gamma \frac{p_0}{\rho_0}}$ is the speed of sound of the solvent \cite{monaghan1994simulating}. Following the weak compression approximation, the value of $c_s$ can be chosen to keep density variations small enough.
The extra stress $\vt \tau_i$ is computed by discretising equation \eqref{eq:K-BKZ_fractional}, as described in the next section. For what concern the boundary conditions, we implemented Newtonian boundary conditions for stress as discussed in \cite{vazquezquesada2012spha}.
\subsubsection{Integral-form implementation of the stress}
The position of each particle is readily accessible in SPH, therefore -to account for periodic boundary conditions- we compute the displacement of a particle compared to the position it had in the past at time $t'=t-n\Delta t$, with $\Delta t$ time step, as $\Delta \boldsymbol{r}_{i,n} = \boldsymbol{r}_i(t) - \boldsymbol{r}'_i(t-n \Delta t)$. It updates as
\begin{eqnarray}
\Delta \boldsymbol{r}_{i,{n+1}} =
\Delta \boldsymbol{r}_{i,n} +
\boldsymbol{r}_i(t+\Delta t) -
\boldsymbol{r}_i(t).
\label{eq:update_displacement}
\end{eqnarray}
The displacement tensor
\begin{eqnarray}
\vb \Delta_i \equiv \qty(\grad \vb r ')_i = -\frac{1}{d_i}\sum_jW'_{ij}\vb e_{ij}\vb r'_{ij}
\end{eqnarray}
is then inverted to compute $\vb E_i$ and $\vt B_i = \vb E_i \vb E^{-T}_i$, so that $\vt \gamma_{[0],i} = \vt 1 - \vt B_i$,
leading to a discretised stress
\begin{equation}
\vt \tau_i = \int_{t_0-T_{save}}^{t_0} M(t-t')h(\gamma)\vt \gamma_{[0],i}(t,t') \dd t' .
\label{stress_calculation}
\end{equation}
which is then efficiently computed over $N_s$ points according to the method described in the supplementary material of \cite{santelli2024smoothed}.
\subsubsection{Numerical parameters}\label{sec:numerical_setup}
%Something introductory
The setup for the numerical experiments consistes of the following. The fluid is confined between two parallel horizontal plates, at a distance $L=7.5$. Periodic boundary conditions are considered for the vertical walls, while no-slip conditions are applied on the horizontal walls. The symmetry of the problems analysed here allow for the simulation to be carried out in a two-dimensional simulation. Thus, the flow domain is set as a square $[0, \, L] \times [0, \, L]$.
The fluid is initially at rest.
The bottom plate is fixed, while the top plate oscillates at a frequency $\omega_0$ and amplitude $\gamma_0=\Delta/L$, i.e. the ratio between the maximum displacement of the upper wall $\Delta$ and $L$. Therefore, for transient shear rate $\gamma(t)$ and strain rate $\gammadot(t)$, we have:
\begin{equation}
\gamma(t) = \gamma_0 \sin(\omega_0 t)
\end{equation}
\begin{equation}
\gammadot(t) = V_{max} \cos(\omega_0 t),
\end{equation}
with $V_{max} = \omega_0 \gamma_0$.
The
number of fluid particles in the system is $N_p = N \times N$: since the domain
is a square, we use the same number of particles in the two dimensions.
Spatial resolution has been previously verified by running simulations at various
$N$, fixing the cut-off radius to $r_c=4L/N$ (see also \cite{santelli2024smoothed}), and we chose $N=60$ to keep the error below $5\%$, resulting in a cut-off radius of $r_c=0.5$.
For the physical parameters chosen for the simulations, we have $\beta_\eta=\eta_p/\eta_{tot}=0.5$, where $\eta_p$ is the polimeric viscosity and $\eta_{tot}=\eta_p+\eta_s$ is the total viscosity, with $\eta_s$ the solvent viscosity. The density $\rho$ is set to unity, and the speed of sound is $c_s=100 \gg V_{max}$, to keep the Mach number $V_{max}/c_s$ small and ensure incompressibility. The characteristic time scale is set upon the the elastic time $\lambda = 1$ (or its closer equivalent when the fractional parameters are outside of the Maxwell regime, see \cite{jaishankar2013power,jaishankar2014fractional}), which ensures that the elastic forces are dominant over the viscous ones and satisfies the series of disequalities $t_{\eta}=\frac{L^2\rho}{\eta_{tot}}<{\lambda}<\frac{2\pi}{\omega_0}$, where $t_{\eta}$ is the viscous time, while $2\pi/\omega_0$ is the period of the oscillations, fixed at a frequency $\omega_0=\lambda^{-1}$.
The time step is set to $\Delta t \approx 10^{-5}\lambda$. For what concerns the parameters of the numerical implementation of the integral model, we have $T_{save}=10\lambda$ and $N_s=100$.
Dimensionless numbers are defined as follows: Reynolds number $Re=\frac{\rho V_{max} L}{\eta_{tot}}$, Weisenberg number $Wi=\frac{\lambda V_{max}}{L}$.
\begin{figure}[tbp]
\centering
\includegraphics[width=0.7\textwidth]{G1G2Sara.png}
\caption{Storage and loss moduli from SAOS: experimental data (\textcolor{blue}{$\bullet$} and \textcolor{red}{$\scalebox{0.6}{$\blacksquare$}$} symbols), fractional K-BKZ model (respectively \textcolor{cyan}{\rule[0.5ex]{0.8em}{0.8pt}} and \textcolor{orange}{\rule[0.5ex]{0.4em}{0.8pt}~\rule[0.5ex]{0.4em}{0.8pt}}) with parameters $\alpha=0.85$, $\beta=0.2$.}
\label{fig:comparison_SAOS}
\end{figure}
\section{Results}\label{sec:results}
In this section we analyse the experimental data obtained from the rheological tests, and use these results to calibrate the parameters of the fractional K-BKZ model. We then compare the experimental data with the numerical results obtained from the SPH simulations.
\subsection{Fitting of experimental data}
\label{sec:fitting_experimental_data}
First, we focus on the choice of the fractional parameters required to capture the experimental data. The initial fractional parameters required calibration are better identified under linear conditions, i.e. small amplitude oscillatory shear (SAOS). Therefore we have performed a series of SAOS tests, where the strain amplitude is kept small enough to ensure that the material behaves linearly and $Re$ is always below $0.01$. The results of these tests are shown in figure \ref{fig:comparison_SAOS}, where we compare the experimental data obtained from runs at $\gamma_0<0.1$ with the numerical predictions of the fractional K-BKZ model under the assumption of small displacement.
We have observed a good agreement for parameters $\alpha=0.85$ and $\beta=0.2$, resulting in model with a fractional behaviour close but distinct from a classical Maxwell model, which would correspond to $\alpha=1$ and $\beta=0$.
\begin{figure}[b]
\centering
\begin{subfigure}{0.48\textwidth}
\centering
\includegraphics[width=\textwidth]{PP_HF955M0_damping_curve_G1G2_largeFont.png}
\caption{Normalised $G'$ and $G''$ (\textcolor{blue}{\rule[0.5ex]{0.8em}{0.8pt}} and \textcolor{orange}{\rule[0.5ex]{0.4em}{0.8pt}~\rule[0.5ex]{0.4em}{0.8pt}}) as a function of strain amplitude $\gamma_0$. The curve is extrapolated from experimental data}
\label{fig:G12_thinning}
\end{subfigure}
\hfill
\begin{subfigure}{0.48\textwidth}
\centering
\includegraphics[width=\textwidth]{PP_HF955M0_damping_curve_GstarG2_largeFont.png}
\caption{Normalised $G^*$ as a function of strain amplitude $\gamma_0$. Symbols for experimental data, solid line for fitted $h(\gamma)$.}
\label{fig:Gstar_thinning}
\end{subfigure}
\caption{Analysis of experimental data at fixed frequency $\omega_0=1$~s$^{-1}$, showing strain thinning.}
\label{fig:experimental_G}
\end{figure}
Then, we focus on the implementation of the non-linear behaviour, where the required function to calibrate is the damping function $h(\gamma)$, defined in equation \eqref{eq:damping_function}. We gathered experimental data at fixed frequency $\omega_0=1$~s$^{-1}$ for two order of magnitudes of values of initial strain $\gamma_0$. In figure \ref{fig:experimental_G}, the values of $G'$, $G''$ and the relaxation modulus $G^*$ are normalised over their respective value at $\gamma_0\to0$. Figure \ref{fig:G12_thinning} clearly shows that the polypropylene we analysed behaves as a strain thinning fluid \cite{hyun2011review}. Therefore, following the approach of \cite{jaishankar2014fractional}, we obtained the fitting parameters for the damping function (see Figure \ref{fig:Gstar_thinning}), which results in $h(\gamma)=(1+0.2\gamma^{1.9})^{-1}$.
Henceforth, the parameters of the numerical scheme are set to the values obtained from the fitting of the experimental data.
\subsection{LAOS analysis}
\label{sec:LAOS_analysis}
The appareance of non-linearity can be observed from multiple points of view. Plotting the elastic and viscous Lissajous curves produces patterns that deviate from ellipses, but rather closed loops, as shown in figure \ref{fig:june1_lissajous}, where the normalised stress $\sigma(\gamma)/\sigma_{max}$ is plotted versus the shear $\gamma(t)$ or the normalised shear $\overline{\gamma(t)}=\gamma(t)/\gamma_{0}$. However, this approach struggles in quantifying the non-linearity, and is especially hard to apply when the non-linear behaviour is just emerging. Indeed, it can be noticed from figure \ref{fig:june1_lissajous_all} that the Lissajous curves for small to medium $\gamma_0$ are almost indistinguishable from the linear case, with the non linearity clearly appearing only for $\gamma_0=5$.
\begin{figure}[bpt]
\centering
\begin{subfigure}{0.48\textwidth}
\centering
\includegraphics[width=\textwidth]{june1_lissajous_all.png}
\caption{Lissajous curves for multiple $\gamma_0$.}
\label{fig:june1_lissajous_all}
\end{subfigure}
\hfill
\begin{subfigure}{0.48\textwidth}
\centering
\includegraphics[width=\textwidth]{june1_lissajous_4.png}
\caption{Lissajous curve for large $\gamma_0$.}
\label{fig:june1_lissajous_4}
\end{subfigure}
\caption{Elastic Lissajous curves from SPH simulations: (a) average curves for $0.2<\gamma_0<5$, (b) details for $\gamma_0=5$. The non-elliptical shape is an indicator of nonlinear viscoelastic behavior.}
\label{fig:june1_lissajous}
\end{figure}
A quantitative analysis of the results requires the application of a Fourier transform to the stress signal, in order to obtain the generalised storage and loss moduli. Following the guidelines outlined in \cite{hyun2011review}, we have chosen to avoid a simple Fast Fourier Transform (FFT) butterfly method, as it may induce shifts in the transformed signal. Therefore, we applied a Direct Fourier Transform (DFT) to the stress signal for the experimental data we have available, which is a more robust method for obtaining the Fourier coefficients. Nevertheless, the amount of data points we have available for the simulated data is often above the recommended limit of DFT, so -after verifying that the results are compatible between the two methods- we have chosen to apply a pseudo-periodic FFT, where, assuming $f(t_1-t_0)$ is the zero-mean signal between two zeros, we construct a longer signal iteratively as $f(t_{n+1}-t_n)=-f(t_{n}-t_{n-1})$, where $t_n$ is the time of the $n$-th zero. The resulting signal is a periodic signal of the chosen lenght, which is then transformed using the FFT method.
Results for SPH simulations of various values of the strain amplitude $\gamma_0$ as a function of $\overline{\omega} = \omega/\omega_1$, with $\omega_1$ the frequency of the first peak, are shown in figure \ref{fig:SPH_fourier_1}, normalised over the value of the first harmonic. As expected, peaks at higher (odd) harmonics appear only for large displacements, and their intensity increases with the strain amplitude.
\begin{figure}[tbp]
\centering
\includegraphics[width=0.8\textwidth]{Fourier_peaks.png}
\caption{Fourier transform of the stress signal for various values of the strain amplitude $\gamma_0$. Higher harmonics appear only for large displacements, and their intensity and amount increase with the strain amplitude.}
\label{fig:SPH_fourier_1}
\end{figure}
\subsection{Comparison with experimental data}
It can be shown that the the storage and loss moduli can be reconstructed from the Fourier transform of the stress signal by adding together the amplitude of the odd harmonics of the Fourier transform. In particular, the first harmonic reproduces the linear response of the material, and the ratio between the peaks of the first and third harmonics ($I_3/I_1$) is a good measure of the non-linearity of the material. We compared the results of the SPH simulations with the experimental data, and we found a good agreement between the two regarding the appareance of non linearity (see figure \ref{fig:fourier_single}), as well as the value of the ratio $I_3/I_1$ over the range of displacements explored. In particular, we found that the value $\gamma_0$ for which $I_3/I_1\sim1\%$, i.e. when the non-linearity becomes non trascurable, is consistently around $1$.
\begin{figure}[tbp]
\centering
\begin{subfigure}{0.48\textwidth}
\centering
\includegraphics[width=\textwidth]{fourier_comparison_july_Wi0.01.png}
\caption{$\gamma_0=0.01$}
\label{fig:small_displacement_comparison}
\end{subfigure}
\hfill
\begin{subfigure}{0.48\textwidth}
\centering
\includegraphics[width=\textwidth]{fourier_comparison_july_Wi1.png}
\caption{$\gamma_0=1.0$}
\label{fig:large_displacement_comparison}
\end{subfigure}
\caption{Comparison of the FFT from experimental data and SPH simulation (respectively \textcolor{cyan}{\rule[0.5ex]{0.4em}{0.8pt}~\rule[0.5ex]{0.4em}{0.8pt}} and \textcolor{orange}{\rule[0.5ex]{0.8em}{0.8pt}}). (a): small displacement; (b): large displacement. Peaks at higher harmonics appear only for large displacements and their intensity matches the experimental data.}
\label{fig:fourier_single}
\end{figure}
From the values of the peaks, multiple non-equivalent definitions for the non-linear quantities of $G'$ and $G''$ can be used. \cite{ewoldt2008new} proposed two local definitions for the elastic modulus, the minimum-strain modulus $G'_M$ and the large-strain modulus $G'_L$, and for the dynamic viscosity: minimum rate viscosity $\eta_M$ and large-rate dynamic viscosity $\eta_L$. These values can be computed from the Fourier coefficients using the following equations:
\begin{align}
G'_M &= \sum_{n\,\text{odd}} n G'_n
&\qquad
G'_L &= \sum_{n\,\text{odd}} (-1)^{(n-1)/2} G'_n \\
\eta_M &= \sum_{n\,\text{odd}} \frac{G''_n}{\omega}
&\qquad
\eta_L &= \sum_{n\,\text{odd}} (-1)^{(n-1)/2} \frac{G''_n}{\omega}
\end{align}
% \begin{align}
% G'_M &= \sum_{n\,\text{odd}} n G'_n \\
% G'_L &= \sum_{n\,\text{odd}} (-1)^{(n-1)/2} G'_n \\
% \eta_M &= \sum_{n\,\text{odd}} G''_n / \omega \\
% \eta_L &= \sum_{n\,\text{odd}} (-1)^{(n-1)/2} G''_n / \omega
% \end{align}
where $G'_n$ and $G''_n$ are the Fourier coefficients of the stress signal. It is clear that for small displacements, where the additional harmonics are negligible, the sums are equivalent to their first term and therefore $G'_M = G'_L$. However, when non-linearity appears, differences between the two definitions can be observed. We have reported our results in figure \ref{fig:GL_GM}, which clearly shows the effects of non-linearity growing with $\gamma_0$ and becoming important when $\gamma_0 \sim 1$.
%\luca{ Maybe add here a comparison with experimental data, for example $G_L^{SPH}$ vs $G_L^{exp}$ and $G_M^{SPH}$ vs $G_M^{exp}$.}
\begin{figure}[tbp]
\centering
\includegraphics[width=0.7\textwidth]{GL_GM_values_log.png}
\caption{Comparison of the minimum-strain modulus $G'_M$ and large-strain modulus $G'_L$ as a function of strain amplitude $\gamma_0$ from SPH simulations. The divergence between $G'_M$ and $G'_L$ at large amplitudes is a clear indicator of nonlinear viscoelastic behavior.}
\label{fig:GL_GM}
\end{figure}
\section{Conclusion}\label{sec:conclusions}
In this work, we have presented a numerical implementation of the fractional K-BKZ model for the simulation of the rheological behaviour of polypropylene-based polymer melts under large amplitude oscillatory shear (LAOS) using the Smoothed Particle Hydrodynamics (SPH) method. The SPH method is a Lagrangian particle-based approach that allows for the simulation of complex fluid flows with large deformations and complex geometries. We have shown that the SPH method is well-suited for implementing integral models, such as the fractional K-BKZ model, which captures the nonlinear viscoelastic behaviour of polymer melts.
Indeed, once the model parameters are fitted to the experimental values in the linear regime, and a dampening function is identified, the SPH simulations recover the occurrence of non-linearity in a consistent way when compared to new experimental data for all the strain amplitudes tested. Therefore, we believe that this implementation has the potential for being used in a wide range of applications, to predict the experimental behaviour of polymer melts under LAOS conditions, and to provide insights into the underlying mechanisms of viscoelasticity in complex fluids, helping the technological development.
It is current under development the analysis of the system when it is not the imposed flow, but rather the geometry, the source of non-linearity, i.e. the case of a polymer melt flowing in a channel with a sudden expansion or contraction. In this case, the flow is not uniform and the shear rate is not constant. Finally, a topic for a future study will be the analysis of a viscous matrix filled with solid particles, which is a common scenario in many industrial applications, such as the processing of polymer composites or the production of polymer-based materials with specific mechanical properties. This would extend the initial promising SPH results presented in \cite{vazquezquesada2019shear} for a viscoelastic suspension with a simple Oldroyd-B matrix.
%The results presented here highlight the effectiveness of combining experimental rheology with advanced numerical modeling. The agreement between SPH simulations and experimental LAOS data demonstrates the potential of fractional integral models for capturing complex viscoelastic behavior. This approach could be extended to other polymeric systems and flow conditions, providing a valuable tool for both fundamental research and industrial applications.
\backmatter
% \bmhead{Supplementary information}
% If your article has accompanying supplementary file/s please state so here.
% Authors reporting data from electrophoretic gels and blots should supply the full unprocessed scans for key as part of their Supplementary information. This may be requested by the editorial team/s if it is missing.
% Please refer to Journal-level guidance for any specific requirements.
\bmhead{Acknowledgements}
This work has been partially funded by the Basque Government through
the ELKARTEK 2022 programme (KAIROS project: grant KK-2022/00052). Financial support through the BERC 2022-2025 program and by the Spanish State Research Agency through BCAM Severo Ochoa Excellence Accreditation CEX2021-001142-S/MICIN/AEI/10.13039/501100011033 and through the projects PID2020-117080RB-C55 (‘Microscopic foundations of soft-matter experiments: computational nano-hydrodynamics’, with acronym ‘Compu-Nano-Hydro) and PID2020-117080RB-C54 (‘‘Coarse-Graining theory
and experimental techniques for multiscale biological systems’’) funded both by AEI – MICIN are also gratefully acknowledged. We also want to thank to Ir. Jint Nijman from Thermo Fisher Scientific, Karlsruhe, for his help in the treatment of the rheometric data for this research.
% \section*{Declarations}
% Some journals require declarations to be submitted in a standardised format. Please check the Instructions for Authors of the journal to which you are submitting to see if you need to complete this section. If yes, your manuscript must contain the following sections under the heading `Declarations':
% \begin{itemize}
% \item Funding
% \item Conflict of interest/Competing interests (check journal-specific guidelines for which heading to use)
% \item Ethics approval and consent to participate
% \item Consent for publication
% \item Data availability
% \item Materials availability
% \item Code availability
% \item Author contribution
% \end{itemize}
% \noindent
% If any of the sections are not relevant to your manuscript, please include the heading and write `Not applicable' for that section.
% %%===================================================%%
% %% For presentation purpose, we have included %%
% %% \bigskip command. Please ignore this. %%
% %%===================================================%%
% \bigskip
% \begin{flushleft}%
% Editorial Policies for:
% \bigskip\noindent
% Springer journals and proceedings: \url{https://www.springer.com/gp/editorial-policies}
% \bigskip\noindent
% Nature Portfolio journals: \url{https://www.nature.com/nature-research/editorial-policies}
% \bigskip\noindent
% \textit{Scientific Reports}: \url{https://www.nature.com/srep/journal-policies/editorial-policies}
% \bigskip\noindent
% BMC journals: \url{https://www.biomedcentral.com/getpublished/editorial-policies}
% \end{flushleft}
% \begin{appendices}
% \section{Section title of first appendix}\label{secA1}
% An appendix contains supplementary information that is not an essential part of the text itself but which may be helpful in providing a more comprehensive understanding of the research problem or it is information that is too cumbersome to be included in the body of the paper.
%%=============================================%%
%% For submissions to Nature Portfolio Journals %%
%% please use the heading ``Extended Data''. %%
%%=============================================%%
%%=============================================================%%
%% Sample for another appendix section %%
%%=============================================================%%
%% \section{Example of another appendix section}\label{secA2}%
%% Appendices may be used for helpful, supporting or essential material that would otherwise
%% clutter, break up or be distracting to the text. Appendices can consist of sections, figures,
%% tables and equations etc.
% \end{appendices}
%%===========================================================================================%%
%% If you are submitting to one of the Nature Portfolio journals, using the eJP submission %%
%% system, please include the references within the manuscript file itself. You may do this %%
%% by copying the reference list from your .bbl file, paste it into the main manuscript .tex %%
%% file, and delete the associated \verb+\bibliography+ commands. %%
%%===========================================================================================%%
%\bibliography{sn-bibliography}% common bib file
% \bibliography{full_lib23}
\bibliography{copy_lib}
%% if required, the content of .bbl file can be included here once bbl is generated
%%\input sn-article.bbl
\end{document}
%% BioMed_Central_Bib_Style_v1.01
\begin{thebibliography}{30}
% BibTex style file: bmc-mathphys.bst (version 2.1), 2014-07-24
\ifx \bisbn \undefined \def \bisbn #1{ISBN #1}\fi
\ifx \binits \undefined \def \binits#1{#1}\fi
\ifx \bauthor \undefined \def \bauthor#1{#1}\fi
\ifx \batitle \undefined \def \batitle#1{#1}\fi
\ifx \bjtitle \undefined \def \bjtitle#1{#1}\fi
\ifx \bvolume \undefined \def \bvolume#1{\textbf{#1}}\fi
\ifx \byear \undefined \def \byear#1{#1}\fi
\ifx \bissue \undefined \def \bissue#1{#1}\fi
\ifx \bfpage \undefined \def \bfpage#1{#1}\fi
\ifx \blpage \undefined \def \blpage #1{#1}\fi
\ifx \burl \undefined \def \burl#1{\textsf{#1}}\fi
\ifx \doiurl \undefined \def \doiurl#1{\url{https://doi.org/#1}}\fi
\ifx \betal \undefined \def \betal{\textit{et al.}}\fi
\ifx \binstitute \undefined \def \binstitute#1{#1}\fi
\ifx \binstitutionaled \undefined \def \binstitutionaled#1{#1}\fi
\ifx \bctitle \undefined \def \bctitle#1{#1}\fi
\ifx \beditor \undefined \def \beditor#1{#1}\fi
\ifx \bpublisher \undefined \def \bpublisher#1{#1}\fi
\ifx \bbtitle \undefined \def \bbtitle#1{#1}\fi
\ifx \bedition \undefined \def \bedition#1{#1}\fi
\ifx \bseriesno \undefined \def \bseriesno#1{#1}\fi
\ifx \blocation \undefined \def \blocation#1{#1}\fi
\ifx \bsertitle \undefined \def \bsertitle#1{#1}\fi
\ifx \bsnm \undefined \def \bsnm#1{#1}\fi
\ifx \bsuffix \undefined \def \bsuffix#1{#1}\fi
\ifx \bparticle \undefined \def \bparticle#1{#1}\fi
\ifx \barticle \undefined \def \barticle#1{#1}\fi
\bibcommenthead
\ifx \bconfdate \undefined \def \bconfdate #1{#1}\fi
\ifx \botherref \undefined \def \botherref #1{#1}\fi
\ifx \url \undefined \def \url#1{\textsf{#1}}\fi
\ifx \bchapter \undefined \def \bchapter#1{#1}\fi
\ifx \bbook \undefined \def \bbook#1{#1}\fi
\ifx \bcomment \undefined \def \bcomment#1{#1}\fi
\ifx \oauthor \undefined \def \oauthor#1{#1}\fi
\ifx \citeauthoryear \undefined \def \citeauthoryear#1{#1}\fi
\ifx \endbibitem \undefined \def \endbibitem {}\fi
\ifx \bconflocation \undefined \def \bconflocation#1{#1}\fi
\ifx \arxivurl \undefined \def \arxivurl#1{\textsf{#1}}\fi
\csname PreBibitemsHook\endcsname
%%% 1
\bibitem[\protect\citeauthoryear{Hyun et~al.}{2011}]{hyun2011review}
\begin{barticle}
\bauthor{\bsnm{Hyun}, \binits{K.}},
\bauthor{\bsnm{Wilhelm}, \binits{M.}},
\bauthor{\bsnm{Klein}, \binits{C.O.}},
\bauthor{\bsnm{Cho}, \binits{K.S.}},
\bauthor{\bsnm{Nam}, \binits{J.G.}},
\bauthor{\bsnm{Ahn}, \binits{K.H.}},
\bauthor{\bsnm{Lee}, \binits{S.J.}},
\bauthor{\bsnm{Ewoldt}, \binits{R.H.}},
\bauthor{\bsnm{McKinley}, \binits{G.H.}}:
\batitle{A review of nonlinear oscillatory shear tests: Analysis and application of large amplitude oscillatory shear (laos)}.
\bjtitle{Progress in Polymer Science}
\bvolume{36}(\bissue{12}),
\bfpage{1697}--\blpage{1753}
(\byear{2011})
\end{barticle}
\endbibitem
%%% 2
\bibitem[\protect\citeauthoryear{Yosick et~al.}{1997}]{yosick1997kinetic}
\begin{barticle}
\bauthor{\bsnm{Yosick}, \binits{J.A.}},
\bauthor{\bsnm{Giacomin}, \binits{A.J.}},
\bauthor{\bsnm{Moldenaers}, \binits{P.}}:
\batitle{A kinetic network model for nonlinear flow behavior of molten plastics in both shear and extension}.
\bjtitle{Journal of non-newtonian fluid mechanics}
\bvolume{70}(\bissue{1-2}),
\bfpage{103}--\blpage{123}
(\byear{1997})
\end{barticle}
\endbibitem
%%% 3
\bibitem[\protect\citeauthoryear{Blackwell and Ewoldt}{2014}]{blackwell2014simple}
\begin{barticle}
\bauthor{\bsnm{Blackwell}, \binits{B.C.}},
\bauthor{\bsnm{Ewoldt}, \binits{R.H.}}:
\batitle{A simple thixotropic–viscoelastic constitutive model produces unique signatures in large-amplitude oscillatory shear (laos)}.
\bjtitle{Journal of Non-Newtonian Fluid Mechanics}
\bvolume{208–209},
\bfpage{27}--\blpage{41}
(\byear{2014})
\end{barticle}
\endbibitem
%%% 4
\bibitem[\protect\citeauthoryear{K{\H{o}}kuti et~al.}{2016}]{koekuti2016exploring}
\begin{barticle}
\bauthor{\bsnm{K{\H{o}}kuti}, \binits{Z.}},
\bauthor{\bsnm{V{\"o}lker-Pop}, \binits{L.}},
\bauthor{\bsnm{Brandst{\"a}tter}, \binits{M.}},
\bauthor{\bsnm{Kokavecz}, \binits{J.}},
\bauthor{\bsnm{Ailer}, \binits{P.}},
\bauthor{\bsnm{Palkovics}, \binits{L.}},
\bauthor{\bsnm{Szab{\'o}}, \binits{G.}},
\bauthor{\bsnm{Czirj{\'a}k}, \binits{A.}}:
\batitle{Exploring the nonlinear viscoelasticity of a high viscosity silicone oil with laos}.
\bjtitle{Applied Rheology}
\bvolume{26}(\bissue{1}),
\bfpage{1}--\blpage{9}
(\byear{2016})
\end{barticle}
\endbibitem
%%% 5
\bibitem[\protect\citeauthoryear{Vargas et~al.}{2023}]{vargas2023multiscale}
\begin{barticle}
\bauthor{\bsnm{Vargas}, \binits{R.O.}},
\bauthor{\bsnm{Gómez-López}, \binits{A.}},
\bauthor{\bsnm{Escandón}, \binits{J.P.}},
\bauthor{\bsnm{Mil-Martínez}, \binits{R.}},
\bauthor{\bsnm{Phillips}, \binits{T.N.}}:
\batitle{Multiscale modeling of complex fluids under saos and laos using a combined fene transient network model}.
\bjtitle{Physica Scripta}
\bvolume{98}(\bissue{2}),
\bfpage{025213}
(\byear{2023})
\end{barticle}
\endbibitem
%%% 6
\bibitem[\protect\citeauthoryear{Bird and Hassager}{1987}]{bird1987dynamics}
\begin{bbook}
\bauthor{\bsnm{Bird}, \binits{R.B.}},
\bauthor{\bsnm{Hassager}, \binits{O.}}:
\bbtitle{Dynamics of Polymeric Liquids, Fluid Mechanics (Dynamics of Polymer Liquids Vol. 1)}
vol. \bseriesno{25},
p. \bfpage{672}.
\bpublisher{Wiley-Interscience}, \blocation{???}
(\byear{1987})
\end{bbook}
\endbibitem
%%% 7
\bibitem[\protect\citeauthoryear{Owens and Phillips}{2002}]{owens2002computational}
\begin{bbook}
\bauthor{\bsnm{Owens}, \binits{R.G.}},
\bauthor{\bsnm{Phillips}, \binits{T.N.}}:
\bbtitle{Computational Rheology},
p. \bfpage{417}.
\bpublisher{Published by Imperial College Press, distribuited by World Scientific Publishing Co.}, \blocation{???}
(\byear{2002})
\end{bbook}
\endbibitem
%%% 8
\bibitem[\protect\citeauthoryear{Jaishankar and McKinley}{2014}]{jaishankar2014fractional}
\begin{barticle}
\bauthor{\bsnm{Jaishankar}, \binits{A.}},
\bauthor{\bsnm{McKinley}, \binits{G.H.}}:
\batitle{A fractional k-{BKZ} constitutive formulation for describing the nonlinear rheology of multiscale complex fluids}.
\bjtitle{Journal of Rheology}
\bvolume{58}(\bissue{6}),
\bfpage{1751}--\blpage{1788}
(\byear{2014})
\end{barticle}
\endbibitem
%%% 9
\bibitem[\protect\citeauthoryear{Santelli et~al.}{2024}]{santelli2024smoothed}
\begin{barticle}
\bauthor{\bsnm{Santelli}, \binits{L.}},
\bauthor{\bsnm{Vázquez-Quesada}, \binits{A.}},
\bauthor{\bsnm{Ellero}, \binits{M.}}:
\batitle{Smoothed particle hydrodynamics simulations of integral multi-mode and fractional viscoelastic models}.
\bjtitle{Journal of Non-Newtonian Fluid Mechanics}
\bvolume{329},
\bfpage{105235}
(\byear{2024})
\end{barticle}
\endbibitem
%%% 10
\bibitem[\protect\citeauthoryear{Gemant}{1938}]{gemant1938fractional}
\begin{barticle}
\bauthor{\bsnm{Gemant}, \binits{A.}}:
\batitle{On fractional differentials}.
\bjtitle{The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science}
\bvolume{25}(\bissue{168}),
\bfpage{540}--\blpage{549}
(\byear{1938})
\end{barticle}
\endbibitem
%%% 11
\bibitem[\protect\citeauthoryear{Blair and Caffyn}{1942}]{blair1942classification}
\begin{barticle}
\bauthor{\bsnm{Blair}, \binits{G.W.S.}},
\bauthor{\bsnm{Caffyn}, \binits{J.}}:
\batitle{The classification of the rheological properties of industrial materials in the light of power-law relations between stress, strain and time}.
\bjtitle{Journal of Scientific Instruments}
\bvolume{19}(\bissue{6}),
\bfpage{88}--\blpage{93}
(\byear{1942})
\end{barticle}
\endbibitem
%%% 12
\bibitem[\protect\citeauthoryear{Jaishankar and McKinley}{2013}]{jaishankar2013power}
\begin{barticle}
\bauthor{\bsnm{Jaishankar}, \binits{A.}},
\bauthor{\bsnm{McKinley}, \binits{G.H.}}:
\batitle{Power-law rheology in the bulk and at the interface: quasi-properties and fractional constitutive equations}.
\bjtitle{Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences}
\bvolume{469}(\bissue{2149}),
\bfpage{20120284}
(\byear{2013})
\end{barticle}
\endbibitem
%%% 13
\bibitem[\protect\citeauthoryear{Yang et~al.}{2010}]{yang2010constitutive}
\begin{barticle}
\bauthor{\bsnm{Yang}, \binits{P.}},
\bauthor{\bsnm{Lam}, \binits{Y.C.}},
\bauthor{\bsnm{Zhu}, \binits{K.-Q.}}:
\batitle{Constitutive equation with fractional derivatives for the generalized {UCM} model}.
\bjtitle{Journal of Non-Newtonian Fluid Mechanics}
\bvolume{165}(\bissue{3-4}),
\bfpage{88}--\blpage{97}
(\byear{2010})
\end{barticle}
\endbibitem
%%% 14
\bibitem[\protect\citeauthoryear{Larson}{1999}]{larson1999structure}
\begin{bbook}
\bauthor{\bsnm{Larson}, \binits{R.G.}}:
\bbtitle{The Structure and Rheology of Complex Fluids},
p. \bfpage{663}.
\bpublisher{Oxford University Press}, \blocation{???}
(\byear{1999})
\end{bbook}
\endbibitem
%%% 15
\bibitem[\protect\citeauthoryear{Viriyayuthakorn and Caswell}{1980}]{viriyayuthakorn1980finite}
\begin{barticle}
\bauthor{\bsnm{Viriyayuthakorn}, \binits{M.}},
\bauthor{\bsnm{Caswell}, \binits{B.}}:
\batitle{Finite element simulation of viscoelastic flow}.
\bjtitle{Journal of Non-Newtonian Fluid Mechanics}
\bvolume{6}(\bissue{3-4}),
\bfpage{245}--\blpage{267}
(\byear{1980})
\end{barticle}
\endbibitem
%%% 16
\bibitem[\protect\citeauthoryear{Rasmussen and Hassager}{1993}]{rasmussen1993simulation}
\begin{barticle}
\bauthor{\bsnm{Rasmussen}, \binits{H.K.}},
\bauthor{\bsnm{Hassager}, \binits{O.}}:
\batitle{Simulation of transient viscoelastic flow}.
\bjtitle{Journal of Non-Newtonian Fluid Mechanics}
\bvolume{46}(\bissue{2–3}),
\bfpage{289}--\blpage{305}
(\byear{1993})
\end{barticle}
\endbibitem
%%% 17
\bibitem[\protect\citeauthoryear{Keunings}{2003}]{keunings2003finite}
\begin{botherref}
\oauthor{\bsnm{Keunings}, \binits{R.}}:
Finite element methods for integral viscoelastic fluids.
Rheology reviews,
167--196
(2003)
\end{botherref}
\endbibitem
%%% 18
\bibitem[\protect\citeauthoryear{Ellero et~al.}{2002}]{ellero2002viscoelastic}
\begin{barticle}
\bauthor{\bsnm{Ellero}, \binits{M.}},
\bauthor{\bsnm{Kr{\"o}ger}, \binits{M.}},
\bauthor{\bsnm{Hess}, \binits{S.}}:
\batitle{Viscoelastic flows studied by smoothed particle dynamics}.
\bjtitle{Journal of Non-Newtonian Fluid Mechanics}
\bvolume{105}(\bissue{1}),
\bfpage{35}--\blpage{51}
(\byear{2002})
\end{barticle}
\endbibitem
%%% 19
\bibitem[\protect\citeauthoryear{Ellero and Tanner}{2005}]{ellero2005sph}
\begin{barticle}
\bauthor{\bsnm{Ellero}, \binits{M.}},
\bauthor{\bsnm{Tanner}, \binits{R.I.}}:
\batitle{{SPH} simulations of transient viscoelastic flows at low reynolds number}.
\bjtitle{Journal of Non-Newtonian Fluid Mechanics}
\bvolume{132}(\bissue{1-3}),
\bfpage{61}--\blpage{72}
(\byear{2005})
\end{barticle}
\endbibitem
%%% 20
\bibitem[\protect\citeauthoryear{V{\'{a}}zquez-Quesada et~al.}{2009}]{vazquezquesada2009smoothed}
\begin{barticle}
\bauthor{\bsnm{V{\'{a}}zquez-Quesada}, \binits{A.}},
\bauthor{\bsnm{Ellero}, \binits{M.}},
\bauthor{\bsnm{Espa{\~{n}}ol}, \binits{P.}}:
\batitle{Smoothed particle hydrodynamic model for viscoelastic fluids with thermal fluctuations}.
\bjtitle{Physical Review E}
\bvolume{79}(\bissue{5}),
\bfpage{056707}
(\byear{2009})
\end{barticle}
\endbibitem
%%% 21
\bibitem[\protect\citeauthoryear{Giacomin et~al.}{2011}]{giacomin2011large}
\begin{barticle}
\bauthor{\bsnm{Giacomin}, \binits{A.J.}},
\bauthor{\bsnm{Bird}, \binits{R.B.}},
\bauthor{\bsnm{Johnson}, \binits{L.M.}},
\bauthor{\bsnm{Mix}, \binits{A.W.}}:
\batitle{Large-amplitude oscillatory shear flow from the corotational maxwell model}.
\bjtitle{Journal of Non-Newtonian Fluid Mechanics}
\bvolume{166}(\bissue{19–20}),
\bfpage{1081}--\blpage{1099}
(\byear{2011})
\end{barticle}
\endbibitem
%%% 22
\bibitem[\protect\citeauthoryear{Rathinaraj et~al.}{2021}]{rathinaraj2021incorporating}
\begin{barticle}
\bauthor{\bsnm{Rathinaraj}, \binits{J.D.J.}},
\bauthor{\bsnm{McKinley}, \binits{G.H.}},
\bauthor{\bsnm{Keshavarz}, \binits{B.}}:
\batitle{Incorporating rheological nonlinearity into fractional calculus descriptions of fractal matter and multi-scale complex fluids}.
\bjtitle{Fractal and Fractional}
\bvolume{5}(\bissue{4}),
\bfpage{174}
(\byear{2021})
\end{barticle}
\endbibitem
%%% 23
\bibitem[\protect\citeauthoryear{Lanyi et~al.}{2019}]{lanyi2019determination}
\begin{botherref}
\oauthor{\bsnm{Lanyi}, \binits{F.J.}},
\oauthor{\bsnm{Wenzke}, \binits{N.}},
\oauthor{\bsnm{Kaschta}, \binits{J.}},
\oauthor{\bsnm{Schubert}, \binits{D.W.}}:
On the determination of the enthalpy of fusion of a‐crystalline isotactic polypropylene using differential scanning calorimetry, x‐ray diffraction, and fourier‐transform infrared spectroscopy: An old story revisited.
Advanced Engineering Materials
\textbf{22}(9)
(2019)
\end{botherref}
\endbibitem
%%% 24
\bibitem[\protect\citeauthoryear{Ellero and Adams}{2010}]{ellero2010sph}
\begin{barticle}
\bauthor{\bsnm{Ellero}, \binits{M.}},
\bauthor{\bsnm{Adams}, \binits{N.A.}}:
\batitle{{SPH} simulations of flow around a periodic array of cylinders confined in a channel}.
\bjtitle{International Journal for Numerical Methods in Engineering}
\bvolume{86}(\bissue{8}),
\bfpage{1027}--\blpage{1040}
(\byear{2010})
\end{barticle}
\endbibitem
%%% 25
\bibitem[\protect\citeauthoryear{Morris et~al.}{1997}]{morris1997modeling}
\begin{barticle}
\bauthor{\bsnm{Morris}, \binits{J.P.}},
\bauthor{\bsnm{Fox}, \binits{P.J.}},
\bauthor{\bsnm{Zhu}, \binits{Y.}}:
\batitle{Modeling low reynolds number incompressible flows using sph}.
\bjtitle{Journal of Computational Physics}
\bvolume{136}(\bissue{1}),
\bfpage{214}--\blpage{226}
(\byear{1997})
\end{barticle}
\endbibitem
%%% 26
\bibitem[\protect\citeauthoryear{Espa{\~{n}}ol and Revenga}{2003}]{espanol2003smoothed}
\begin{barticle}
\bauthor{\bsnm{Espa{\~{n}}ol}, \binits{P.}},
\bauthor{\bsnm{Revenga}, \binits{M.}}:
\batitle{Smoothed dissipative particle dynamics}.
\bjtitle{Physical Review E}
\bvolume{67}(\bissue{2}),
\bfpage{026705}
(\byear{2003})
\end{barticle}
\endbibitem
%%% 27
\bibitem[\protect\citeauthoryear{Monaghan}{1994}]{monaghan1994simulating}
\begin{barticle}
\bauthor{\bsnm{Monaghan}, \binits{J.J.}}:
\batitle{Simulating free surface flows with {SPH}}.
\bjtitle{Journal of Computational Physics}
\bvolume{110}(\bissue{2}),
\bfpage{399}--\blpage{406}
(\byear{1994})
\end{barticle}
\endbibitem
%%% 28
\bibitem[\protect\citeauthoryear{V{\'{a}}zquez-Quesada and Ellero}{2012}]{vazquezquesada2012spha}
\begin{barticle}
\bauthor{\bsnm{V{\'{a}}zquez-Quesada}, \binits{A.}},
\bauthor{\bsnm{Ellero}, \binits{M.}}:
\batitle{{SPH} simulations of a viscoelastic flow around a periodic array of cylinders confined in a channel}.
\bjtitle{Journal of Non-Newtonian Fluid Mechanics}
\bvolume{167-168},
\bfpage{1}--\blpage{8}
(\byear{2012})
\end{barticle}
\endbibitem
%%% 29
\bibitem[\protect\citeauthoryear{Ewoldt et~al.}{2008}]{ewoldt2008new}
\begin{barticle}
\bauthor{\bsnm{Ewoldt}, \binits{R.H.}},
\bauthor{\bsnm{Hosoi}, \binits{A.}},
\bauthor{\bsnm{McKinley}, \binits{G.H.}}:
\batitle{New measures for characterizing nonlinear viscoelasticity in large amplitude oscillatory shear}.
\bjtitle{Journal of Rheology}
\bvolume{52}(\bissue{6}),
\bfpage{1427}--\blpage{1458}
(\byear{2008})
\end{barticle}
\endbibitem
%%% 30
\bibitem[\protect\citeauthoryear{V{\'{a}}zquez-Quesada et~al.}{2019}]{vazquezquesada2019shear}
\begin{barticle}
\bauthor{\bsnm{V{\'{a}}zquez-Quesada}, \binits{A.}},
\bauthor{\bsnm{Espa{\~{n}}ol}, \binits{P.}},
\bauthor{\bsnm{Tanner}, \binits{R.I.}},
\bauthor{\bsnm{Ellero}, \binits{M.}}:
\batitle{Shear thickening of a non-colloidal suspension with a viscoelastic matrix}.
\bjtitle{Journal of Fluid Mechanics}
\bvolume{880},
\bfpage{1070}--\blpage{1094}
(\byear{2019})
\end{barticle}
\endbibitem
\end{thebibliography}