The Selection-Stitch Model: Resolving the
Hubble Tension via Holographic Phase
Transitions in a Polycrystalline Vacuum
Raghu Kulkarni
∗
Independent Researcher, Calabasas, CA
February 25, 2026
Abstract
The persistent discrepancy between early-universe (H
0
≈ 67.4 km/s/Mpc) [6]
and late-universe (H
0
≈ 73.0 km/s/Mpc) [5] measurements of the Hubble constant
strongly suggests a physical breakdown in the standard ΛCDM model. We pro-
pose that this tension arises not from missing particles or elusive dark fluids, but
from a geometric phase transition within the vacuum structure itself. In compan-
ion papers, we established that the vacuum is a discrete, holographically generated
Face-Centered Cubic (FCC, K = 12) tensor network governed by the Kepler kissing
number theorem [16]. Built upon the lattice gauge theory requirement of mini-
mal gauge-invariant triangular loops [3] and a strict zero-parameter e
−3
≈ 4.98%
topological tunneling amplitude [4], the vacuum naturally crystallizes into a poly-
crystalline manifold [1]. Consequently, the early-universe expansion rate is not
merely a fitted parameter, but the bare holographic baseline locked to this satu-
rated 2D boundary, flawlessly predicting the scalar spectral index (n
s
≈ 0.9646)
[2]. Here, we demonstrate that the accumulated Regge deficit strain of the expand-
ing lattice forces the structure to fracture, forming topological grain boundaries
(matter) and exposing pristine macroscopic gaps (cosmic voids). This localized
topological activation shifts the effective coordination number of the vacuum unit
cell from a shielded, surface-limited state (ν = 12) to an exposed, porous state
(ν = 13). This void-induced phase transition predicts an intrinsic expansion boost
of 13/12 ≈ 8.3%, natively amplifying the 67.4 baseline to exactly 73.02 km/s/Mpc.
This single zero-parameter geometric ratio naturally resolves the Hubble tension
and predicts measurable environmental anisotropies for local H
0
surveys without
modifying standard FLRW cosmology in the early universe.
Contents
1 INTRODUCTION 3
2 FORMALISM: THE CRYSTALLINE VACUUM 3
2.1 Hamiltonian Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.2 Defining the “Stitch”: Entanglement as Geometry . . . . . . . . . . . . . . 4
2.3 The Cosmological Dictionary . . . . . . . . . . . . . . . . . . . . . . . . . . 4
∗
raghu@idrive.com
1
3 THERMODYNAMIC DERIVATION OF GROUND STATE 5
4 KINEMATIC VERIFICATION AND THE HOLOGRAPHIC LATTICE 5
5 DERIVATION OF THE HUBBLE TENSION 6
5.1 Phase I: Shielded (Early Universe & The Holographic Baseline) . . . . . . 6
5.2 Phase II: Exposed (Late Universe & Void Activation) . . . . . . . . . . . . 7
5.3 The Boost Calculation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
6 COMPATIBILITY WITH STANDARD COSMOLOGY 8
6.1 Recovery of the FLRW Continuum . . . . . . . . . . . . . . . . . . . . . . 8
6.2 Measurable Changes to Late-Time Cosmology . . . . . . . . . . . . . . . . 8
7 EQUATION OF STATE 9
8 CONCLUSION 9
A Thermodynamic Emergence of the FCC Lattice 10
B Monte Carlo Validation of Topological Activation 10
2
1 INTRODUCTION
Precision cosmology has arrived at a fascinating impasse. A persistent 5σ tension exists
between the Hubble constant derived from the early universe via the Planck Cosmic
Microwave Background (CMB) measurements [6] and the late universe measured via local
SH0ES distance ladders [5]. While standard theoretical solutions frequently invoke early
dark energy or novel particle species to patch the discrepancy [7], we propose looking at
the foundational fabric of space itself.
What if the vacuum undergoes a structural phase transition, much like systems in
condensed matter physics? The principle of Universality in statistical mechanics dictates
that macroscopic phase transitions depend primarily on symmetry and dimensionality,
rather than microscopic, fine-tuned details. In physics, we routinely observe fluid-to-
lattice transitions, such as the Wigner crystallization of electrons [8], Abrikosov vortex
lattices in superconductors [9], and Coulomb crystals trapped in ion confinement [10]. It
is a natural theoretical extension to ask whether the entanglement network of the vacuum
itself behaves similarly [11].
In this manuscript, we extend the Selection-Stitch Model (SSM). As established in
recent foundational work, spacetime is not a smooth, pre-existing container, but emerges
thermodynamically as a discrete tensor network of quantum entanglement [2]. We demon-
strate that:
1. The vacuum naturally relaxes into a Face-Centered Cubic (FCC) lattice to minimize
free energy and maximize its information capacity.
2. Standard FLRW cosmology and the precise CMB baseline measurements are per-
fectly preserved at high cosmic densities.
3. The formation of massive cosmic voids at late times triggers a topological phase
transition in the lattice connectivity from ν = 12 to ν = 13, amplifying the expan-
sion rate by exactly 13/12.
2 FORMALISM: THE CRYSTALLINE VACUUM
2.1 Hamiltonian Definition
Rather than assuming a continuous background, we define the pre-geometric vacuum as
the ground state of a stabilizer Hamiltonian acting on a dynamic graph G(V, E)—conceptually
akin to Kitaev’s Toric Code models [12]. The Hamiltonian is given by:
H = −J
e
X
v
A
v
− J
m
X
p
B
p
(1)
where A
v
are vertex operators that maximize local connectivity (entanglement), and B
p
are plaquette operators enforcing geometric constraints (metric flatness and quantum
error correction) [13]. The coupling constant J
e
represents the energetic benefit of a
unitary “stitch” (an entanglement link), while J
m
enforces the 1/
√
3L exclusion radius
[1], preventing the nodes from collapsing into a singularity.
3
2.2 Defining the “Stitch”: Entanglement as Geometry
In the Selection-Stitch Model, a link is not a physical tether, but a fundamental unit of
quantum information. We formally define a Stitch as a single, discrete unit of quantum
entanglement (e.g., a Bell pair) shared between two localized vacuum domains. Follow-
ing the Ryu-Takayanagi geometric interpretation of entanglement [14], spatial proximity
is strictly an emergent property of quantum correlation. Two nodes in the tensor net-
work are considered adjacent “neighbors” (separated by exactly one fundamental lattice
unit, L) precisely because they share maximal entanglement entropy. Consequently, the
dynamical act of “stitching” is the generation of validated entanglement links between
states. A broken stitch corresponds to a localized loss of correlation, which manifests
macroscopically as a metric gap or cosmic void.
2.3 The Cosmological Dictionary
To translate these discrete lattice statistics into the continuous Hubble parameter H(t)
observed by astrophysicists, we derive the dependence of the macroscopic expansion rate
on the underlying lattice topology.
1. Volume Scaling: Because 3D bulk volume is fundamentally a holographic pro-
jection of 2D bounding surfaces [1], we assume that the cosmological volume V (t) is
extensive with the total number of bulk nodes N(t). Therefore, the scale factor a(t)
scales as a(t) ∝ N(t)
1/3
.
2. The Expansion Rate: The standard Hubble parameter is defined as H = ˙a/a.
Differentiating our scaling relation yields:
H(t) =
1
3
˙
N(t)
N(t)
(2)
The node nucleation rate
˙
N is governed strictly by the 2D holographic surface availability.
New nodes can only nucleate at the exposed faces of the unit cells on the vacuum boundary
(or along internal void boundaries):
˙
N ∝ N
surf ace
× ν × Γ
0
∝ N
2/3
ν (3)
where ν is the number of active, unshielded nucleation channels per unit cell.
3. The Topological Link: Substituting
˙
N into the Hubble equation reveals the core
scaling relation:
H(t) ∝
N(t)
2/3
· ν
N(t)
= ν · N(t)
−1/3
(4)
While H(t) naturally decays over time as N
−1/3
(which is entirely consistent with standard
decelerating expansion), for any fixed epoch (a fixed N), the expansion rate is linearly
proportional to the available topological degrees of freedom ν:
H ∝ ν (for fixed N) (5)
Thus, if the vacuum undergoes a topological phase transition that alters the integer value
of ν, it will manifest observationally as an instantaneous step-change in the Hubble pa-
rameter.
4
3 THERMODYNAMIC DERIVATION OF GROUND
STATE
Why does the vacuum select a specific geometry? As derived in Appendix A, we utilize
a Free Energy functional F (k) governing the vacuum node connectivity k:
F (k) = −αk + βe
k−12
− T S
sym
(6)
where α = J
e
(the vertex coupling from Eq. 1) and β corresponds to the plaquette stiffness
J
m
. This functional perfectly balances three competing physical drives:
• Entanglement Drive (−αk): The system seeks to maximize connectivity to in-
crease its computational entropy [15].
• Metric Wall (βe
k−12
): The geometry enforces the rigid Kepler Conjecture limit,
preventing more than 12 neighbors from packing around a central sphere in 3D space
[16].
• Isotropy Selection (−T S
sym
): The functional naturally selects the isotropic FCC
lattice (O
h
symmetry) over the anisotropic HCP lattice to minimize rotational en-
tropy penalties.
We emphasize that the FCC geometry is not phenomenologically fine-tuned to fit cosmo-
logical data. It is an independent mathematical consequence of the Kepler Conjecture
and the requirement for spatial isotropy. The resulting 13/12 ratio explored below is a
rigid downstream mathematical constraint, not an input parameter.
4 KINEMATIC VERIFICATION AND THE HOLO-
GRAPHIC LATTICE
To ensure this geometry is physically realizable via constructive operations (rather than
just existing as a thermodynamic assumption), this discrete manifold was explicitly ver-
ified via computational modeling. As rigorously established in our companion computa-
tional analysis [1], the SSM kinematics naturally saturate at K = 12 without jamming.
Crucially, this saturation is achieved with zero free parameters. By strictly enforc-
ing the lattice gauge theory requirement that the minimal physical entanglement structure
is a closed triangular loop [3], the lattice natively expands as 2D hexagonal sheets. The
generation of 3D volume occurs solely via a quantum topological tunneling event from a
1D to a 0D solution manifold, governed by the exact amplitude P = e
−3
≈ 4.98% [4].
Operating under this parameter-free thermodynamic paradigm, and bounded by the
exact 1/
√
3L geometric exclusion limit, the network reliably compiles into a polycrys-
talline Face-Centered Cubic bulk.
5
Figure 1: Constructive Verification of Vacuum Geometry. A 3D visualization of
the tensor network simulation (N = 5, 000) demonstrating polycrystalline FCC crystal-
lization, detailed thoroughly in [1]. The yellow core indicates nodes that have successfully
reached the Kepler Saturation limit (K = 12). The stability of this core proves that
the SSM operators natively form a coherent spatial manifold without jamming into an
amorphous glass state.
Because perfect tetrahedral tiling of flat 3D space is geometrically impossible—leaving
a Regge deficit angle of δ ≈ 7.36
◦
—the lattice cannot form a single monolithic crystal.
Instead, the simulation yields a modal saturation (∼ 41.1% at N = 5000), with the
remaining nodes locked in 1D and 2D topological grain boundaries (K < 12). The
confirmation that the local geometric ground state is explicitly Cuboctahedral (K = 12)
bounded by defect grain boundaries provides the rigid, integer-based topology required
to evaluate macroscopic expansion metrics.
5 DERIVATION OF THE HUBBLE TENSION
We now translate this lattice topology into observable cosmology using the dictionary
derived in Eq. 5.
5.1 Phase I: Shielded (Early Universe & The Holographic Base-
line)
In the high-density regime characterizing the Early Universe and the CMB era, the un-
derlying lattice is fully saturated. A central space-time node is deeply “Shielded” by its
12 immediate neighbors in the FCC unit cell (the Cuboctahedron). The number of active
directions available for the nucleation of new space is strictly limited to the existing 2D
6
external lattice vectors:
ν
early
= 12 (7)
As explicitly derived in [2], this precise 2D holographic geometry dictates the fractional
curvature perturbation of the early universe, flawlessly predicting the CMB scalar spectral
index (n
s
≈ 0.9646). Therefore, the Planck measurement of 67.4 km/s/Mpc is not a
tension-generating anomaly; it is the mathematically true, base-level expansion rate of
the fully shielded 2D holographic grid.
5.2 Phase II: Exposed (Late Universe & Void Activation)
In the late universe, non-linear matter clustering creates massive Cosmic Voids. In the
SSM framework, the origin of these voids and the matter that bounds them is fundamen-
tally topological. Because of the accumulated geometric strain of the δ ≈ 7.36
◦
Regge
deficit angle, the continuous K = 12 early-universe lattice must eventually fracture.
This fracturing separates the vacuum into pristine crystalline domains (macroscopic
Cosmic Voids) bounded by topological grain boundaries of frozen K < 12 defects (which
manifest macroscopically as ordinary matter and large-scale cosmic structure). As the
universe expands and these pristine K = 12 domains dilate, matter is geometrically
swept to the void perimeters.
Inside a void, matter density functionally vanishes, exposing the bare geometric lattice
and creating macroscopic vacancies in the grid. A vacancy inherently breaks the perfect
O
h
symmetry of the unit cell.
Figure 2: The Topological Phase Transition. Left: In the early universe (Shielded
Phase), the vacuum unit cell is fully coordinated (ν = 12), restricting expansion to the
surface boundaries. Right: In the late universe (Exposed Phase), cosmic voids create
localized metric vacancies. A vacancy breaks the symmetry, activating a 13th “bulk”
vector (red arrow) into the void, driving a localized boost in the expansion rate.
Crucially, the vacancy removes one occupied matter site, but it does not destroy the
underlying geometric vector; instead, it converts the central reference node from a passive
structural anchor into an active nucleation site along the direction of the void. The net
effect is the original 12 surface vectors plus 1 newly activated bulk channel:
ν
late
= 12(Surface) + 1(Bulk/Void) = 13 (8)
7
5.3 The Boost Calculation
The ratio of the late-to-early expansion rates is simply the ratio of these active topological
channels:
H
late
H
early
=
ν
late
ν
early
=
13
12
≈ 1.0833 (9)
Applying this precise geometric boost to the theoretically anchored Planck CMB mea-
surement yields the local expansion rate:
H
pred
= 67.4 ×
13
12
≈ 73.02 km/s/Mpc (10)
This first-principles calculation natively matches the SH0ES local measurement (73.04 ±
1.04) without utilizing any continuous fitting parameters or altering standard early-
universe dynamics.
6 COMPATIBILITY WITH STANDARD COSMOL-
OGY
A radical proposal must first do no harm to established physics. A critical requirement
for any proposed vacuum framework is the pristine preservation of the highly success-
ful standard cosmological epochs (radiation and matter domination) prior to late-time
acceleration.
6.1 Recovery of the FLRW Continuum
During the radiation-dominated and early matter-dominated eras, the universe is charac-
terized by a relatively uniform, high-density distribution of mass-energy. In the discrete
geometric framework, particles are localized topological defects entangled within the vac-
uum grid. Because the background lattice remains fully saturated and structurally uni-
form (K = 12, Shielded Phase) under this density, the discrete tensor network perfectly
coarse-grains into a smooth, continuous manifold.
Consequently, the macroscopic thermodynamic behavior of the lattice exactly recov-
ers the standard Friedmann-Lemaˆıtre-Robertson-Walker (FLRW) metric. The expansion
rate scales exactly as a(t) ∝ t
1/2
(radiation) and a(t) ∝ t
2/3
(matter) because the topo-
logical multiplier (ν = 12) remains an unbroken, universal constant across all space. The
model is therefore mathematically indistinguishable from standard ΛCDM during the
epoch of Recombination, preserving all precision measurements of the Cosmic Microwave
Background (CMB).
6.2 Measurable Changes to Late-Time Cosmology
The framework’s departure from standard ΛCDM is surgical: it occurs exclusively at late
times, triggered by the onset of non-linear structure formation. As matter collapses into
dense filaments and superclusters, it evacuates vast regions of space, generating macro-
scopic cosmic voids. It is only within these matter-empty voids that the bare vacuum
lattice is exposed, triggering the localized ν = 12 → 13 topological phase transition.
This physical mechanism yields specific, highly measurable deviations from standard
cosmology:
8
1. Environmental H
0
Anisotropy: Because the expansion boost is physically teth-
ered to the formation of voids, the model predicts that local H
0
measurements will
exhibit a strict environmental dependence. Distance ladders calibrated through
underdense regions (voids) will natively yield H
0
≈ 73.0 km/s/Mpc, while measure-
ments calibrated along high-density structural filaments (where the vacuum remains
shielded) will remain closer to the bare 67.4 km/s/Mpc background. This provides
a clear, measurable resolution to the directional anisotropies increasingly observed
in modern local universe surveys.
2. Step-Function Transition: Unlike the Cosmological Constant (Λ), which exerts
a smooth, homogeneous acceleration across all space, this topological activation
predicts a discrete “knee” in the effective expansion rate corresponding exactly to
the epoch of widespread void percolation (z ∼ 1 to 2).
7 EQUATION OF STATE
Finally, we explore how this geometric relaxation mimics standard dark energy equations
of state w(z). We define the effective dark energy density ρ
DE
as proportional to the
vacuum channel density, which naturally scales with the void fraction η(z):
ρ
DE
(z) = ρ
vac
(1 + η(z)) (11)
Based on the standard ansatz that void volume scales inversely with matter density,
η(z) ∝ ρ
−1
m
∝ (1 + z)
−3
. Substituting this into the continuity equation for dark energy
yields:
w = −1 −
1
3
d ln ρ
DE
d ln a
(12)
Given that η ≪ 1 at the onset of the transition, we approximate ln(1+η) ≈ η. To maintain
a rigorous, parameter-free model, we fix the coupling constant to the exact topological
boost derived earlier (ϵ = 1/12). This yields:
w(z) ≈ −1 +
1
12
(1 + z)
−3
(13)
This mathematically predicts a “thawing” equation of state (w > −1), a dynamic evolu-
tion highly consistent with recent high-precision data from the DESI collaboration [17].
8 CONCLUSION
The Selection-Stitch Model provides an intuitive, self-contained, first-principles derivation
of vacuum geometry. As established in foundational computational work [1, 2], an FCC
lattice bounded by a 2D holographic surface is the natural kinematic ground state of
the quantum tensor network. This uniform K = 12 lattice beautifully recovers standard
FLRW cosmology during the radiation and early matter epochs.
With the fundamental topological defects of this network naturally predicting early-
universe inflation and the CMB spectral index (n
s
≈ 0.9646), this manuscript resolves
the Hubble Tension by identifying the specific topological mechanism governing late-time
acceleration. A localized phase transition of the lattice (12 → 13)—driven exclusively by
9
cosmic void formation resulting from Regge deficit strain—natively amplifies the funda-
mental 67.4 expansion baseline to exactly 73.02 km/s/Mpc. This parameter-free geometric
ratio natively resolves the Hubble Tension without invoking early dark energy, dark fluids,
or missing particles, while yielding falsifiable predictions for environmental anisotropies
in future deep-sky surveys.
A Thermodynamic Emergence of the FCC Lattice
In this appendix, we derive the vacuum energy landscape directly from fundamental
principles. We posit that the cooling vacuum evolves to maximize its Information Storage
Capacity (Entropy) subject to geometric volumetric constraints.
1. The Entanglement Drive: We define the Unitary Stitch as a single unit of
quantum entanglement. Since entropy S is additive, the system natively seeks to maximize
the number of links k. The energy benefit scales linearly:
E
bind
= −J
e
· k = −αk (14)
2. The Metric Wall: The Kepler Conjecture dictates that the maximum local
density in 3D space is limited to the kissing number K = 12. As k → 12, the available
phase space volume vanishes. We mathematically model this repulsive potential as an
asymptotic wall:
E
repel
= βe
k−12
(15)
We acknowledge that the exponential form is an ansatz; however, any function diverging
at k = 12 (such as a power law) would yield the exact same geometric ground state
selection.
3. Symmetry Selection: Two distinct lattices satisfy the maximum density ρ =
π/
√
18: FCC and HCP. However, HCP possesses a preferred axis (D
6h
), which creates an
entropic penalty in an isotropic plasma. The FCC lattice (O
h
) is strictly isotropic. The
free energy functional minimizes this structural penalty:
F
select
= −T · S
sym
(16)
Combining these terms yields the elegant functional utilized in Section III.
B Monte Carlo Validation of Topological Activation
To computationally validate the mean-field shift from 12 to 13, we performed a Monte
Carlo simulation on a 15
3
grid. The full open-source script utilized to compute this
transition is publicly available at: https://github.com/raghu91302/ssmtheory/blob/
main/ssm-repair-simulation.py
We initialized a fully saturated lattice (K = 12) and introduced random geometric
voids.
• Method: We removed nodes with probability P
void
and dynamically recalculated
the effective degrees of freedom for the remaining nodes based on the “Exposed
Phase” definition (Eq. 8).
• Result: As the void fraction exceeded the standard percolation threshold (∼ 20%),
the mean-field active site count (ν) shifted smoothly and definitively from 12.00 to
12.99.
10
This provides strong computational evidence that the topological transition is not a math-
ematical artifact, but a robust, inevitable feature of void formation in FCC lattices.
References
[1] R. Kulkarni, “Constructive Verification of K = 12 Lattice Saturation: Exploring
Kinematic Consistency in the Selection-Stitch Model,” Zenodo (2026). (Under Peer
Review). DOI: 10.5281/zenodo.18294925.
[2] R. Kulkarni, “Geometric Phase Transitions in a Discrete Vacuum: Deriving Cosmic
Flatness, Inflation, and Reheating from Tensor Network Topology,” Zenodo (2026).
(Under Peer Review). DOI: 10.5281/zenodo.18727238.
[3] K. G. Wilson, “Confinement of quarks,” Phys. Rev. D 10, 2445-2459 (1974).
[4] S. Coleman, “The fate of the false vacuum. 1. Semiclassical theory,” Phys. Rev. D
15, 2929-2936 (1977).
[5] A. G. Riess et al., “A Comprehensive Measurement of the Local Value of the Hubble
Constant,” arXiv:2112.04510 (2022).
[6] Planck Collaboration, “Planck 2018 results. VI. Cosmological parameters,” A&A
641, A6 (2020).
[7] M. Kamionkowski and A. G. Riess, “The Hubble Tension and Early Dark Energy,”
Ann. Rev. Nucl. Part. Sci. 73 (2023).
[8] E. Wigner, “On the Interaction of Electrons in Metals,” Phys. Rev. 46, 1002 (1934).
[9] A. A. Abrikosov, “On the Magnetic Properties of Superconductors of the Second
Group,” Sov. Phys. JETP 5, 1174 (1957).
[10] G. Birkl et al., “Multiple-Shell Structures of Laser-Cooled Ions,” Nature 357, 310
(1992).
[11] M. Van Raamsdonk, “Building up spacetime with quantum entanglement,” Gen. Rel.
Grav. 42, 2323-2329 (2010).
[12] A. Kitaev, “Fault-tolerant quantum computation by anyons,” Annals of Physics 303,
2 (2003).
[13] A. Almheiri, X. Dong, and D. Harlow, “Bulk locality and quantum error correction
in AdS/CFT,” JHEP 2015, 163 (2015).
[14] S. Ryu and T. Takayanagi, “Holographic derivation of entanglement entropy from
AdS/CFT,” Phys. Rev. Lett. 96, 181602 (2006).
[15] S. Lloyd, “Computational capacity of the universe,” Phys. Rev. Lett. 88, 237901
(2002).
[16] T. C. Hales, “A proof of the Kepler conjecture,” Annals of Mathematics 162, 1065
(2005).
11
[17] DESI Collaboration, “DESI 2024 VI: Cosmological Constraints,” arXiv:2404.03002
(2024).
12
