\documentclass[11pt]{article}

\usepackage[margin=1in]{geometry}
\usepackage{amsmath}
\usepackage{amssymb}
\usepackage{graphicx}
\usepackage{booktabs}
\usepackage{tikz}
\usepackage{pgfplots}
\pgfplotsset{compat=1.18}
\usepackage[colorlinks=true, linkcolor=black, citecolor=black, urlcolor=blue]{hyperref}

% --- skin palette: pecorino gold, roasted-pepper brown, Mozzarella-Phase rust, smooth teal
\definecolor{cpgold}{HTML}{A8720F}
\definecolor{cppepper}{HTML}{4A3B28}
\definecolor{cpclump}{HTML}{A1382A}
\definecolor{cpsmooth}{HTML}{1F7A6C}
\definecolor{cpref}{HTML}{8A7960}

\title{The Phase Transition Hiding in Your Cacio e Pepe}
\author{The Science Journaling Club}
\date{Field Note --- Paper Analysis --- Soft Matter Physics}

\begin{document}
\maketitle

\begin{abstract}
Cacio e pepe is three ingredients --- pecorino, black pepper, pasta water --- and it fails
constantly, seizing into a grainy, stringy mess. In April 2025, eight physicists gave that
failure a name and a boundary. Writing in \textit{Physics of Fluids}, Bartolucci and colleagues
show that the clumped state is a genuine thermodynamic phase, the \textbf{Mozzarella Phase},
separated from a smooth one by a starch concentration of almost exactly \textbf{1\% relative to
the cheese's mass} \cite{bartolucci2025}. Below that line the sauce clumps once it passes roughly
\textbf{65\,\textdegree C}; above it, starch holds the cheese proteins apart to 95\,\textdegree C.
The transition is a lower critical solution temperature, sitting slightly above
\textbf{60\,\textdegree C} at a protein mass fraction of \textbf{$\phi \approx 0.134$}, and it runs
backward from kitchen folklore: it is heating the mixture too far, not letting it cool, that
breaks it. We work through how the phase diagram was mapped and do \textbf{our own arithmetic} on
the recommended recipe: 5\,g of starch against 200\,g of cheese is \textbf{2.5\% by mass}, about
\textbf{1\tfrac{7}{8} teaspoons} by a standard kitchen conversion, and it puts the dish itself at a
protein fraction of \textbf{$\phi \approx 0.154$} --- close enough to the paper's most clump-prone
concentration that the 1\% line is not a formality.
\end{abstract}

%============================================================================
\section{A Phase Diagram for Dinner}

Cacio e pepe should be the easiest dish in the Roman repertoire. Cook pasta. Reserve some of the
starchy water. Melt Pecorino Romano into that water off the heat, whisk in black pepper, toss with
pasta, serve. Three ingredients, one pan, no oven. And it fails so reliably that Italian cooking
forums have entire threads devoted to nothing else: instead of a glossy cream the cheese seizes
into rubbery threads and grit, sitting in a thin, separated liquid. Restaurant kitchens have worked
out folk fixes over generations --- more starch, less heat, add the cheese slower --- without anyone
writing down why any of it works.

In January 2025, eight physicists posted a preprint that treats that failure as exactly what a
condensed-matter physicist would call it: a phase transition in a multi-component mixture, with a
real phase diagram, a real critical point, and a controllable order parameter. It ran in
\textit{Physics of Fluids} in April, authored by Giacomo Bartolucci, Daniel Maria Busiello, Matteo
Ciarchi, Alberto Corticelli, Ivan Di Terlizzi, Fabrizio Olmeda, Davide Revignas and Vincenzo Maria
Schimmenti \cite{bartolucci2025}. Several of them do this kind of physics for a living on
considerably less delicious systems --- phase separation in polymer mixtures and the physics of
biomolecular condensates inside living cells \cite{bartolucci2023pnas} --- and the paper reads like
people who noticed their day job sitting in their dinner. Cooking has drawn serious physicists'
attention before, from pasta's own fracture mechanics \cite{ball2010} to a broader review of fluid
dynamics across the kitchen \cite{mathijssen2023}; this paper's contribution is to bring the same
seriousness specifically to a sauce.

\textit{A note on scope. This is a field note on someone else's published measurements. The club
ran no burner and boiled no pasta for this piece. The phase diagram, the temperatures, and the
recipe below are the paper's; the unit conversions and the arithmetic checking the recipe against
the phase diagram are ours, and are marked as such throughout.}

%============================================================================
\section{What Actually Goes Wrong in the Bowl}

Pecorino Romano is not a simple solid that melts. It is an emulsion holding two populations of milk
protein in suspension: \textbf{casein}, which sits in the cheese as large, roughly spherical
clusters called micelles, and \textbf{whey proteins}, smaller globular proteins folded into a
compact shape that keeps their sticky interior hidden. When the cheese is intact and cold, both
populations mostly keep to themselves.

Grate that cheese into hot water and both populations start to change. Whey proteins denature ---
the folded shape opens up, exposing hydrophobic patches that used to be tucked inside --- and once
denatured, they are unusually eager to stick to themselves, to each other, and to the outside of
casein micelles \cite{anema2003,anema2021}. The casein micelles, meanwhile, aggregate on their own
as they warm \cite{nicolai2021}. Push enough heat into the system and individually-dispersed
protein starts finding itself more attracted to more protein than to the water around it.
Aggregates form, grow, and merge into visible clumps big enough to trap fat and leave separated
liquid behind. That is what a home cook watches happen when the sauce ``breaks,'' and it is what
the paper's authors, tongue only partly in cheek, name the \textbf{Mozzarella Phase}: a texture that
would be exactly right in a different dish, and a disaster in this one.

%============================================================================
\section{The Backward Physics of a Broken Sauce}

Here is the part of the paper that overturns a piece of kitchen folklore. Home cooks who have had a
sauce break tend to blame it on the sauce cooling unevenly or on cheese hitting cold pasta, and the
usual advice is to keep things off the heat and warm. The measured physics runs the other way: with
no starch present, cheese mixed with hot water forms large clumps once the mixture crosses about
\textbf{65\,\textdegree C} --- clumping is triggered by getting \textit{too hot}, not by cooling
down \cite{bartolucci2025}.

The formal name for this shape of phase behavior is a \textbf{lower critical solution temperature}
(LCST). Most mixing intuition runs the opposite way: oil and vinegar separate cold and blend warm;
salt dissolves better hot than cold. An LCST system is the reverse. Below the critical temperature
the components mix into one homogeneous phase; above it, the mixture splits into a protein-rich
phase and a protein-poor phase. Whey proteins denaturing on heating supplies the switch: cold,
undenatured whey stays folded and sociable toward water; hot, denatured whey flips to a state that
prefers sticking to other protein instead \cite{bartolucci2025,anema2003}. Push a cacio e pepe past
that switch and the physics does not care how carefully the emulsion was built up to that point.

%============================================================================
\section{Turning a Ruined Sauce Into Data}

Knowing the direction of the effect is not the same as mapping it. To build an actual phase
diagram, the team needed a reproducible way to measure ``how clumped'' a sample was, across many
combinations of ingredients and temperatures.

Their method: mix cheese, water, and a chosen starch concentration, spread the mixture thinly
across a transparent Petri dish, and backlight it from below. Aggregated protein blocks light; a
smooth sauce lets it through evenly. Ramp the temperature from \textbf{50\,\textdegree C to
95\,\textdegree C in 5\,\textdegree C steps}, over roughly \textbf{40 minutes} per run,
photographing the dish at every step. The photographs are then normalized to greyscale, smoothed
with a Gaussian filter ($\sigma = 1$ pixel), binarized with a quantile threshold swept between 0.05
and 0.25 to isolate dark aggregate regions, cleaned up with watershed segmentation, small-object
removal, dilation and hole-filling, and finally approximated as ellipses. The major axis length of
the largest fitted ellipse becomes the order parameter --- growing from zero in a smooth sample to
the width of the dish in a fully clumped one \cite{bartolucci2025}. Every ramp was run at least
twice.

\begin{figure}[htbp]
\centering
\begin{tikzpicture}[line width=0.9pt, font=\footnotesize]
  \draw[->] (0,0) -- (5.4,0) node[right] {starch \%};
  \draw[->] (0,0) -- (0,4.4) node[above] {temperature};
  \foreach \x/\lbl in {0/0, 1.55/1, 3.1/2, 5.0/4} \draw (\x,0.05) -- (\x,-0.05) node[below] {\lbl\%};
  \foreach \y/\lbl in {0/50, 1.3/65, 2.6/80, 4.0/95} \draw (0.05,\y) -- (-0.05,\y) node[left] {\lbl};
  % smooth region shading
  \fill[cpsmooth!12] (1.55,0) rectangle (4.6,4.2);
  \fill[cpref!22] (4.6,0) rectangle (5.4,4.2);
  % mozzarella region under the rising boundary
  \fill[cpclump!18] plot[smooth, tension=0.7] coordinates {(0,0)(0,1.55)(0.55,2.8)(1.1,3.9)(1.55,4.2)} -- (0,4.2) -- cycle;
  \draw[cpclump, very thick] plot[smooth, tension=0.7] coordinates {(0,1.55)(0.55,2.8)(1.1,3.9)(1.55,4.2)};
  \draw[cpgold, dashed, thick] (1.55,0) -- (1.55,4.2) node[above] {1\% threshold};
  \draw[cpref, dashed] (4.6,0) -- (4.6,4.2);
  \fill[cpclump] (0,1.55) circle (1.6pt) node[right] {\scriptsize 65\textdegree C onset, no starch};
  \fill[cpgold] (3.1,2.05) circle (2pt) node[right] {\scriptsize recipe: 2.5\% starch};
  \node[cpclump] at (0.65,2.2) {\scriptsize Mozzarella};
  \node[cpsmooth] at (2.9,3.6) {\scriptsize Smooth};
  \node[cpref] at (5.0,3.6) {\scriptsize stiff};
\end{tikzpicture}
\caption{The paper's central result, redrawn schematically. Below 1\% starch the sauce enters the
Mozzarella Phase once heated past a boundary anchored at about 65\,\textdegree C with no starch at
all; at or above 1\% the mixture stays smooth through 95\,\textdegree C; past roughly 4\% a
different, non-clumping failure (an overly stiff sauce) appears. The 8 starch levels actually
tested were 0, 0.5, 1, 1.5, 2, 2.5, 3 and 4\% \cite{bartolucci2025}; the connecting curve is our
schematic of the boundary's reported shape, not a redigitization of the measured images.}
\end{figure}

%============================================================================
\section{The Line at One Percent}

With a working order parameter, the team ran two systematic sweeps. First: hold cheese and water at
equal mass (1:1) and vary starch across \textbf{0, 0.5, 1, 1.5, 2, 2.5, 3 and 4\%} of the cheese's
mass, at every temperature step from 50 to 95\,\textdegree C. Second: fix starch at 1\% and vary the
cheese-to-water ratio instead \cite{bartolucci2025}.

The result is close to a bright line. Below 1\% starch, the sauce enters the Mozzarella Phase once
heated past its critical temperature. At or above 1\%, the same cheese and water stay smooth all
the way to 95\,\textdegree C. Push starch past roughly 4\% and a new failure mode appears: not
clumping, but a sauce thick and gluey with gelatinized starch. The paper's practical sweet spot is
\textbf{2--3\% starch}, comfortably inside the smooth region and short of the stiff one
\cite{bartolucci2025}.

%============================================================================
\section{The Free Energy Underneath the Boundary}

The boundary is grounded in the physics of polymer solutions. Mix a solute made of large molecules
(proteins, starch) into a solvent (water), and the classic way to write the free energy of that
mixture per unit volume is the Flory--Huggins form, built independently by Paul Flory and Maurice
Huggins in 1942 \cite{flory1942,huggins1942}:
\begin{equation}
\frac{\Delta G_{\text{mix}}}{k_B T} \;\approx\; \frac{\phi}{n}\ln\phi \;+\; (1-\phi)\ln(1-\phi)
\;+\; \chi\,\phi(1-\phi),
\end{equation}
where $\phi$ is the protein mass fraction, $n$ captures how large the protein-bearing species is
relative to the solvent, and $\chi$ is the interaction parameter describing how much solute and
solvent would rather not mix. The first two terms are entropy: mixing is favored simply because
there are more ways to arrange a mixed system. The last term is energy, and it can win: raise $\chi$
enough and the free energy develops two minima instead of one, meaning the system prefers splitting
into a protein-rich phase and a protein-poor phase --- exactly the clumping the paper measures.

What makes this an LCST system rather than the more familiar kind is that heating \textit{increases}
$\chi$ here. The paper attributes that rise to the biochemistry above: warming denatures whey
protein and drives casein micelles to aggregate, both making protein-protein contact more
energetically favorable relative to protein-water contact as temperature climbs
\cite{bartolucci2025}. The size parameter $n$ is also allowed to grow with temperature, interpreted
as starch forming larger starch--protein complexes as the system heats \cite{bartolucci2025,liu2023}.

%============================================================================
\section{Where the Line Sits: the Binodal and Its Minimum}

The second sweep --- fixed 1\% starch, varying cheese-to-water ratio --- traces the actual phase
boundary (the \textbf{binodal}) in terms of protein concentration. Plotted as protein mass fraction
$\phi$ against temperature, the boundary is a parabola opening upward, with its minimum --- the
easiest point at which to trigger phase separation --- at a temperature \textbf{slightly above
60\,\textdegree C} and $\phi \approx 0.134$, obtained from cheese and water at a plain 1:1 ratio by
mass \cite{bartolucci2025}. To convert cheese quantity into protein fraction, the authors used the
nutritional label of Pecorino Romano DOP: \textbf{27\,g of protein per 100\,g of cheese}
\cite{bartolucci2025}. At 1:1 cheese to water, half the mixture's mass is cheese, and
$0.5 \times 0.27 \approx 0.134$ --- exactly the minimum the binodal traces.

\begin{figure}[htbp]
\centering
\begin{tikzpicture}[line width=0.9pt, font=\footnotesize]
  \draw[->] (0,0) -- (5.6,0) node[right] {$\phi$ (protein mass fraction)};
  \draw[->] (0,0) -- (0,4.4) node[above] {temperature};
  \foreach \x/\lbl in {0.8/0.05, 2.0/0.10, 3.0/0.134, 4.2/0.20, 5.3/0.25} \draw (\x,0.05) -- (\x,-0.05) node[below] {\lbl};
  \foreach \y/\lbl in {0/50, 1.4/60, 2.6/70, 3.8/80} \draw (0.05,\y) -- (-0.05,\y) node[left] {\lbl};
  \fill[cpclump!14] plot[smooth, tension=0.75] coordinates {(0.9,4.2)(1.7,1.9)(3.0,1.15)(4.3,1.9)(5.1,4.2)} -- (5.1,4.2) -- (0.9,4.2) -- cycle;
  \draw[cpclump, very thick] plot[smooth, tension=0.75] coordinates {(0.9,4.2)(1.7,1.9)(3.0,1.15)(4.3,1.9)(5.1,4.2)};
  \fill[cpgold] (3.0,1.15) circle (2pt) node[above right] {\scriptsize LCST min: $\phi\approx0.134$, $>$60\textdegree C};
  \draw[cpref, dashed] (3.0,0) -- (3.0,4.2);
  \node[cpclump] at (3.0,3.4) {\scriptsize two phases (clumped)};
  \node[cpsmooth] at (3.0,0.5) {\scriptsize one phase (smooth)};
  \draw[cpgold, thick] (3.6,1.65) circle (2.4pt) node[right] {\scriptsize recipe (no starch): $\phi\approx0.154$};
\end{tikzpicture}
\caption{The binodal in the coordinates the paper uses for the starch-free system. Below the curve,
one homogeneous phase; above it, two coexisting phases --- the definition of a lower critical
solution temperature. The minimum sits at $\phi \approx 0.134$, slightly above 60\,\textdegree C
\cite{bartolucci2025}. The open point is our own placement of the actual recipe's cheese-and-water
ratio (before starch), worked out in Section~9.}
\end{figure}

%============================================================================
\section{A Sharper Fix: Trisodium Citrate}

Trisodium citrate --- the salt used to keep processed cheese smooth, and a common trick in
modernist cheese-sauce cooking \cite{bertsch2019} --- gets the same treatment: swept across
concentrations relative to cheese mass through identical temperature ramps.

The comparison is instructive. Starch defends the sauce gradually: as concentration rises through
the danger zone, clump size falls off over a spread of a percent or two. Citrate behaves more like a
switch: at \textbf{1\% citrate the Mozzarella Phase still forms} at high enough temperature; at
\textbf{2\% the sauce is fully stabilized}, and 3\% confirms it --- a transition the authors
describe as sharp, explicitly contrasted with starch's gradual one \cite{bartolucci2025}. The likely
reason is mechanistic: citrate is a small ion that can chelate the calcium holding casein micelles
together and directly disrupt protein-protein bonding, while starch's protection is closer to a
steric and complexation effect --- crowding aggregating proteins apart and possibly binding some of
the denatured whey directly \cite{bartolucci2025,liu2023}. A direct chemical disruption should show
up as a sharper boundary than a physical crowding effect, and that is what the data show. The
tradeoff is that citrate blunts the cheese's flavor more than starch does.

%============================================================================
\section{From Phase Diagram to Recipe}

The paper closes with a recipe built directly on the measured boundary, for two servings
\cite{bartolucci2025}:

\begin{center}
\begin{tabular}{@{}ll@{}}
\toprule
Ingredient & Quantity \\
\midrule
Pasta & 300\,g tonnarelli (spaghetti or rigatoni work too) \\
Cheese & 200\,g Pecorino Romano DOP, finely grated \\
Starch & 5\,g potato or corn starch \\
Water & 50\,g to dissolve the starch, plus 100\,g after (150\,g total) \\
Pepper & whole black peppercorns, toasted, then cracked \\
\bottomrule
\end{tabular}
\end{center}

Dissolve the starch in the first 50\,g of water and heat gently until it gelatinizes --- the swelling
and loss of crystalline order in the starch granules that thickens the mixture \cite{jenkins1998}.
Cool the gel,
stir in the remaining 100\,g of water, then blend with the grated cheese into a smooth cream. Toast
the pepper and fold it in. Cook the pasta separately, cool it about a minute, combine off the direct
heat, finish with reserved pasta water as needed, and serve at 80--90\,\textdegree C. The
alternative: replace the starch with 5\,g of trisodium citrate dissolved in the full 150\,g of
water --- a sharper fix, at some cost to flavor \cite{bartolucci2025}.

%============================================================================
\section{The Club's Arithmetic: Checking the Recipe Against the Diagram}

The paper states its threshold as a percentage and hands over a recipe in grams. We did the
conversion ourselves --- our own arithmetic, run on the paper's own reported numbers, appearing
nowhere in the paper itself.

\textbf{Where the recipe sits on the starch axis.} 5\,g of starch against 200\,g of cheese:
\begin{equation}
\frac{5\text{ g starch}}{200\text{ g cheese}} = 0.025 = 2.5\%.
\end{equation}
That lands inside the paper's ``ideal'' 2--3\% window, and is $2.5\times$ the bare 1\% survival
threshold --- not a knife-edge margin, but not a large one either.

\textbf{Where that is in a measuring spoon.} Using the standard kitchen conversion of roughly 8\,g
of cornstarch per tablespoon (so about 2.67\,g per teaspoon, a common kitchen reference figure and
not a number from this paper at all):
\begin{equation}
5\text{ g} \;\div\; 2.67\text{ g/tsp} \;\approx\; 1.87\text{ teaspoons}.
\end{equation}
Call it a scant two teaspoons. The difference between a smooth cacio e pepe and the Mozzarella
Phase, in this recipe, is roughly the difference between two level teaspoons of starch and one.

\textbf{Where the recipe sits on the protein axis.} This is the calculation we think matters most,
because it connects the recipe directly to the binodal of Section~7. Using the same definition of
$\phi$ the paper uses, and the same 27\,g-protein-per-100\,g figure for Pecorino Romano
\cite{bartolucci2025}, the recipe's own cheese and water (starch excluded from this axis exactly as
the paper excludes it) give:
\begin{equation}
\phi_{\text{recipe}} = \frac{200\text{ g} \times 0.27}{200\text{ g} + 150\text{ g}}
= \frac{54\text{ g}}{350\text{ g}} \approx 0.154.
\end{equation}
The actual dish sits at a protein concentration close to --- and just past --- the single most
clump-prone point on the entire phase diagram. Less water than the recipe calls for pushes $\phi$
even higher, deeper into the danger zone; more water pulls it toward either edge of the parabola,
where the two-phase region is narrower. We think this is the real reason the 1\% starch line matters
as much as the paper's data say it does: the traditional dish is not made at a safely dilute protein
concentration. It is made almost exactly where clumping is easiest.

\textbf{A consistency check.} The paper separately describes its citrate alternative as ``2.5\% by
cheese mass,'' with the same 5\,g against the same 200\,g of cheese \cite{bartolucci2025} --- exactly
the 2.5\% our division above produces for starch too. Reassuring, when arithmetic done two different
ways on two different ingredients lands on the same number.

%============================================================================
\section{Why This Is Actually Legitimate Physics}

It would be easy to read all of this as a joke dressed in equations --- physicists cannot resist
measuring things, so they measured dinner. We think that reading misses the point. A phase
transition does not care what industry supplies its ingredients. Casein micelles aggregating under
heat, whey proteins denaturing and turning sticky, a polymer additive sterically stabilizing a
colloid against that aggregation, an LCST separating one phase from two --- every piece of that
vocabulary belongs equally to a dairy science paper, a materials science paper on protein gels, or a
kitchen. The authors are not stretching physics to cover a novelty topic; the physics was already
there, unexamined, because nobody had run the experiment on a system this mundane.

There is a second reason the choice of system is not incidental. Several of the paper's authors work,
in their non-culinary research, directly on liquid--liquid phase separation and biomolecular
condensates --- the same physics governing how cells organize their interior without membranes,
currently one of the more active areas of biophysics \cite{brangwynne2009,hyman2014}. Giacomo
Bartolucci co-authored a 2023 PNAS paper on cyclic phase separation as a mechanism for selecting
molecular sequences during the origin of life \cite{bartolucci2023pnas}, work that lives in exactly
the same theoretical framework --- free energy landscapes, interaction parameters, binodal curves ---
used here to explain a broken sauce. The mathematics Flory and Huggins built in 1942 to describe
rubber and polymer melts organizes a startlingly wide range of real systems, from protein droplets
inside a living cell down to a pot on a stove, and demonstrating that range on a system every reader
has personally ruined is, if anything, better pedagogy than doing it on a system nobody outside a lab
has touched.

None of that erases the humor. ``Mozzarella Phase'' is a deliberately funny name, chosen by people
clearly enjoying themselves. Funny and rigorous are not opposites here; the joke is precisely that
the rigor was warranted all along.

%============================================================================
\section{What We Would Want to See Next}

A few open threads, in the order we would chase them. \textbf{Other cheeses}: the whole diagram
rests on Pecorino Romano's specific protein content and casein-to-whey ratio; Parmigiano Reggiano,
aged differently, would likely move the boundary, and nothing described here tests that.
\textbf{Other starches}: potato and corn starch gelatinize at different temperatures and
viscosities, and whether the 1\% line holds identically for both is a natural next measurement.
\textbf{The mechanism behind the size parameter}: the growth of $n$ with temperature is inferred
from a fit rather than measured directly; light scattering on the actual mixture could turn that
inference into a demonstrated mechanism. \textbf{Whether the margin generalizes}: our
$\phi \approx 0.154$ calculation is specific to this exact recipe and this cheese's protein content.
Whether traditional recipes across cuisines that emulsify hard cheese into hot liquid tend to
cluster near the most demanding point of their own phase diagrams, or whether that is peculiar to
cacio e pepe, is the more general and more interesting question.

%============================================================================
\begin{thebibliography}{99}

\bibitem{bartolucci2025}
Bartolucci, G., Busiello, D.~M., Ciarchi, M., Corticelli, A., Di Terlizzi, I., Olmeda, F.,
Revignas, D. \& Schimmenti, V.~M. (2025). Phase behavior of Cacio e Pepe sauce.
\textit{Physics of Fluids} 37, 044122. doi:10.1063/5.0255841

\bibitem{ball2010}
Ball, P. (2010). Pasta physics. \textit{Nature Materials} 9, 539. doi:10.1038/nmat2793

\bibitem{mathijssen2023}
Mathijssen, A.~J.~T.~M., Lisicki, M., Prakash, V.~N. \& Mossige, E.~J.~L. (2023). Culinary fluid
mechanics and other currents in food science. \textit{Reviews of Modern Physics} 95, 025004.
doi:10.1103/RevModPhys.95.025004

\bibitem{bertsch2019}
Bertsch, P., Savorani, L. \& Fischer, P. (2019). Rheology of Swiss Cheese Fondue. \textit{ACS
Omega} 4, 1103--1109. doi:10.1021/acsomega.8b02424

\bibitem{anema2003}
Anema, S.~G. \& Li, Y. (2003). Association of denatured whey proteins with casein micelles in
heated reconstituted skim milk and its effect on casein micelle size. \textit{Journal of Dairy
Research} 70, 73--83. doi:10.1017/S0022029902005903

\bibitem{anema2021}
Anema, S.~G. (2021). Heat-induced changes in caseins and casein micelles, including interactions
with denatured whey proteins. \textit{International Dairy Journal} 122, 105136.
doi:10.1016/j.idairyj.2021.105136

\bibitem{nicolai2021}
Nicolai, T. \& Chassenieux, C. (2021). Heat-induced gelation of casein micelles. \textit{Food
Hydrocolloids} 118, 106755. doi:10.1016/j.foodhyd.2021.106755

\bibitem{liu2023}
Liu, Y., Tan, Z., Xu, X., Liu, J., Zhu, B. \& Dong, X. (2023). The formation of rice starch/casein
complex by hydrothermal treatment and its use in stabilizing high internal phase emulsions.
\textit{Food Hydrocolloids} 144, 108995. doi:10.1016/j.foodhyd.2023.108995

\bibitem{jenkins1998}
Jenkins, P.~J. \& Donald, A.~M. (1998). Gelatinisation of starch: a combined SAXS/WAXS/DSC and
SANS study. \textit{Carbohydrate Research} 308, 133--147. doi:10.1016/S0008-6215(98)00079-2

\bibitem{flory1942}
Flory, P.~J. (1942). Thermodynamics of High Polymer Solutions. \textit{The Journal of Chemical
Physics} 10, 51--61. doi:10.1063/1.1723621

\bibitem{huggins1942}
Huggins, M.~L. (1942). Thermodynamic properties of solutions of long-chain compounds.
\textit{Annals of the New York Academy of Sciences} 43, 1--32.
doi:10.1111/j.1749-6632.1942.tb47940.x

\bibitem{brangwynne2009}
Brangwynne, C.~P., Eckmann, C.~R., Courson, D.~S., Rybarska, A., Hoege, C., Gharakhani, J.,
J\"ulicher, F. \& Hyman, A.~A. (2009). Germline P granules are liquid droplets that localize by
controlled dissolution/condensation. \textit{Science} 324, 1729--1732.
doi:10.1126/science.1172046

\bibitem{hyman2014}
Hyman, A.~A., Weber, C.~A. \& J\"ulicher, F. (2014). Liquid-Liquid Phase Separation in Biology.
\textit{Annual Review of Cell and Developmental Biology} 30, 39--58.
doi:10.1146/annurev-cellbio-100913-013325

\bibitem{bartolucci2023pnas}
Bartolucci, G., Cala\c{c}a Serr\~ao, A., Schwintek, P., K\"uhnlein, A., Rana, Y., Janto, P., Hofer,
D., Mast, C.~B., Braun, D. \& Weber, C.~A. (2023). Sequence self-selection by cyclic phase
separation. \textit{Proceedings of the National Academy of Sciences} 120, e2218876120.
doi:10.1073/pnas.2218876120

\end{thebibliography}

\end{document}