Curbing PBHs with PTAs

This paper demonstrates that pulsar timing array observations of scalar-induced gravitational waves impose stringent constraints on the abundance of primordial black holes, particularly for stellar-mass objects, though these limits are significantly relaxed by positive non-Gaussianities or alternative formation theories like the peaks formalism.

The Gist

The Big Picture: Listening to the Universe's Echoes

Imagine the early universe as a giant, chaotic drum. Sometimes, this drum gets hit so hard that it creates massive, dense clumps of matter that collapse into Primordial Black Holes (PBHs). These are black holes born in the first split second of the universe, long before stars existed.

For a long time, scientists wondered: How many of these baby black holes are hiding in our universe today? Could they make up all the "dark matter" (the invisible stuff holding galaxies together)?

This paper uses a new tool to answer that question: Pulsar Timing Arrays (PTAs). Think of PTAs as a cosmic drumbeat detector. They listen to the rhythmic pulses of dead stars (pulsars) across the galaxy. If gravitational waves (ripples in space-time) pass by, they slightly mess up the timing of these pulses. Recently, the NANOGrav collaboration (a team of astronomers) heard a low-frequency "hum" in the data. They aren't sure what caused it yet, but it's a real signal.

The Connection: Black Holes Make Noise

Here is the catch: You can't just create a bunch of massive PBHs without making a lot of noise.

1. The Cause: To make a PBH, you need a huge "bump" in the density of the early universe.
2. The Effect: That same huge bump doesn't just create a black hole; it also creates a ripple in space-time called a Scalar-Induced Gravitational Wave (SIGW).

Think of it like this: If you try to build a massive sandcastle (a PBH) on a beach, you have to move a lot of sand. That movement creates a wave (the SIGW). You cannot have the castle without the wave.

The paper argues that if there were too many PBHs, the "waves" they created would be so loud that they would drown out the signal NANOGrav is actually hearing. Since the signal NANOGrav hears is relatively quiet, there is a strict limit on how many PBHs could have been made.

The Main Findings

1. The "Stellar Mass" Limit
The authors calculated that for black holes with masses similar to our Sun (stellar mass), the universe is very quiet. This means only a tiny fraction of dark matter can be made of these black holes. If there were too many, the gravitational wave "hum" would be much louder than what we observe.

2. The Shape of the "Bump" Matters
The universe's density bumps aren't all the same shape. Some are sharp spikes (narrow), and some are wide hills (broad).

• Narrow spikes: These create very specific, sharp waves. The PTA data puts tight limits on these.
• Wide hills: These create a broader range of waves. The limits are different but still restrictive.

3. The "Non-Gaussian" Twist (The Secret Sauce)
In physics, we usually assume random fluctuations are "Gaussian" (like a perfect bell curve). But the early universe might have been "Non-Gaussian" (skewed or lopsided).

• The Analogy: Imagine rolling dice. A "Gaussian" roll gives you average numbers most of the time. A "Non-Gaussian" roll might be rigged to give you lots of 6s or lots of 1s.
• The Result: The paper found that if the universe was "rigged" with positive non-Gaussianity (a specific kind of skew), it becomes much easier to make black holes without making too much noise.
  • If this "rigging" is strong enough (specifically, if a parameter called $f_{NL}$ is greater than 5), the PTA constraints basically disappear. We could have a lot more black holes than we thought!
  • However, if the "rigging" is negative, the limits get even stricter, ruling out almost all black holes in certain mass ranges.

4. The "Peaks" vs. "Threshold" Debate
There are two different ways scientists calculate how many black holes form:

• Threshold Statistics: You count how many bumps are "tall enough" to collapse. This method says: "Nope, very few black holes allowed."
• Theory of Peaks: You look at the highest peaks in the landscape. This method is more lenient and says: "Okay, maybe a few more are allowed."
  The paper shows that depending on which math you use, the answer changes by a few orders of magnitude. This highlights that we still have some theoretical uncertainty.

The Supermassive Black Hole Question

We know huge black holes (Supermassive Black Holes or SMBHs) sit in the centers of galaxies. How did they get so big so fast? Maybe they started as "seeds" (smaller PBHs).

The paper checks if the PTA data allows for these seeds.

• The Problem: The scales needed to make SMBH seeds are usually constrained by other data (like the Cosmic Microwave Background).
• The Hope: The authors found that if the universe had very strong "cubic" non-Gaussianity (a more complex type of skewing, parameter $g_{NL}$), it might be possible to create enough massive seeds to grow into SMBHs without violating the PTA rules. However, they note that we don't have a concrete model yet that explains how the universe would create such strong skewing.

Summary in One Sentence

By listening to the gravitational wave "hum" detected by Pulsar Timing Arrays, this paper concludes that there probably aren't many Sun-sized primordial black holes (unless the early universe had some very specific, weird statistical quirks), but there might still be room for them to act as seeds for the giant black holes we see today.

Technical Summary

Technical Summary: Curbing PBHs with PTAs

Problem Statement
Primordial Black Holes (PBHs) are a compelling candidate for dark matter and potential seeds for Supermassive Black Holes (SMBHs). Their formation requires a significant enhancement of the primordial curvature power spectrum ($P_\zeta$) at scales smaller than those probed by the Cosmic Microwave Background (CMB). However, such large curvature perturbations inevitably generate a secondary signal: Scalar-Induced Gravitational Waves (SIGWs). Recent data from Pulsar Timing Arrays (PTAs), specifically the NANOGrav 15-year dataset, have detected a stochastic gravitational wave background (SGWB) with a Hellings-Downs pattern. This paper investigates the tension between the PBH abundance required to explain dark matter or SMBH seeds and the upper limits on the SIGW amplitude imposed by PTA observations. A critical aspect of this investigation is the role of primordial non-Gaussianities (NGs), which can significantly alter both the PBH formation probability and the resulting SIGW spectrum.

Methodology
The authors employ a multi-step theoretical framework to derive constraints on PBH abundance:

1. Power Spectrum Models: The study considers three benchmark shapes for the enhanced curvature power spectrum: a Log-Normal (LN) distribution and two Broken Power-Law (BPL) models with varying widths.
2. Non-Gaussianity (NG) Formalism: The primordial curvature perturbation $\zeta$ is modeled using a local expansion including quadratic ($f_{NL}$) and cubic ($g_{NL}$) terms. The authors analyze both the impact of NGs on the probability density function (PDF) of density fluctuations and their effect on the induced GW spectrum.
3. PBH Abundance Calculation: Two distinct formalisms are used to estimate the PBH mass fraction ($f_{PBH}$):
   • Threshold Statistics: Based on the compaction function and the probability of density fluctuations exceeding a critical threshold ($C_{th}$). This approach accounts for the non-linear relation between curvature and density perturbations.
   • Theory of Peaks: Based on the statistics of local maxima in the curvature field. The authors note that including NGs in this formalism is technically challenging, so this analysis is primarily restricted to the Gaussian limit for comparison.
4. SIGW Computation: The SIGW background is calculated at second order in perturbation theory. The authors derive the GW energy density spectrum ($\Omega_{GW}$), incorporating the effects of $f_{NL}$ on the spectrum's amplitude and shape.
5. PTA Constraints: The theoretical SIGW predictions are compared against the 95% confidence level upper bounds from the NANOGrav 15-year data across 14 frequency bins. The analysis accounts for spectral resolution effects using different window functions (Dirac, Sinc, Top-Hat).
6. SMBH Seed Viability: The paper examines whether large positive NGs can allow for a sufficient abundance of massive PBHs ($10^3 - 10^6 M_\odot$) to seed SMBHs without violating constraints from CMB spectral distortions (FIRAS).

Key Contributions and Results

• Constraints on Stellar Mass PBHs:

  • Under the assumption of Gaussian fluctuations (or NGs arising solely from non-linear density-curvature relations), PTA observations place stringent upper bounds on the amplitude of the curvature power spectrum.
  • Using threshold statistics, the authors find that only a small fraction of dark matter can consist of stellar-mass PBHs. For narrow spectra (LN, BPL1), constraints of $f_{PBH} \lesssim 10^{-5}$ apply to the mass range $0.1 - 100 M_\odot$. For broader spectra (BPL2), constraints apply to heavier masses ($5 - 500 M_\odot$).
  • Using the theory of peaks, the constraints on $f_{PBH}$ are relaxed by several orders of magnitude compared to threshold statistics, highlighting significant theoretical uncertainty in PBH abundance estimates.

• Impact of Primordial Non-Gaussianities:

  • Positive NGs ($f_{NL} > 0$): These have a dual effect. They increase the SIGW amplitude (tightening the bound on the power spectrum amplitude $A$) but, more importantly, they exponentially enhance the probability of large density fluctuations. Consequently, a smaller power spectrum amplitude is needed to produce the same PBH abundance. The net result is that positive NGs relax the constraints on PBH abundance. For $f_{NL} \gtrsim 5$, the PTA constraints on PBH abundance in the stellar mass range are effectively eliminated for broad spectra.
  • Negative NGs ($f_{NL} < 0$): These suppress the tail of the PDF, requiring a larger power spectrum amplitude to achieve a given PBH abundance. This leads to stricter constraints on $f_{PBH}$. The most stringent constraints occur around $f_{NL} \approx -2$.
  • Ultra-Slow Roll (USR) Models: In specific USR inflation scenarios where NGs are generated by the spectral slope, the induced NGs are relatively small ($f_{NL} \sim 1$). In these cases, PTA constraints remain effective, particularly for wider power spectra where NG-induced suppression is weaker.

• SMBH Seeds and Large NGs:

  • The paper addresses the "SMBH seed problem," where standard Gaussian models fail to produce enough massive PBHs without violating CMB spectral distortion limits.
  • The authors demonstrate that large cubic non-Gaussianities ($g_{NL} \sim 10^5$) can overcome this hurdle, allowing for a sufficient abundance of $10^3 - 10^6 M_\odot$ PBHs to seed SMBHs while remaining consistent with FIRAS $\mu$-distortion limits. However, they note that explicit models generating such large $g_{NL}$ values are currently lacking in the literature.

• Theoretical Uncertainties:

  • The choice of abundance formalism (Threshold vs. Peaks) introduces order-of-magnitude differences in the derived constraints.
  • The treatment of the collapse threshold $C_{th}$ in the presence of strong NGs carries uncertainties, though the authors estimate that perturbative corrections to $C_{th}$ would not drastically alter the main conclusions.

Significance and Claims
The paper claims to provide updated, conservative bounds on the PBH abundance derived from the latest PTA data, explicitly incorporating the effects of primordial non-Gaussianities. The primary significance lies in demonstrating that:

1. PTA data is a powerful probe of early-universe physics, capable of ruling out PBHs as the dominant component of dark matter in the stellar mass range under standard Gaussian assumptions.
2. Non-Gaussianities are a critical degeneracy: Positive NGs can "curb" the constraints, allowing for a higher PBH abundance than previously thought possible under Gaussian assumptions, while negative NGs tighten them.
3. Theoretical robustness is limited: The large discrepancy between constraints derived from threshold statistics and peak theory underscores the need for more precise calculations of PBH formation thresholds and abundance in non-Gaussian regimes.
4. SMBH seeding is possible but model-dependent: While large NGs offer a theoretical pathway to seed SMBHs with PBHs, the lack of concrete models producing such large NGs remains a barrier to confirming this scenario.

The authors conclude that while PTA observations significantly constrain the parameter space for PBH dark matter, the final verdict on PBH abundance remains subject to theoretical uncertainties regarding the collapse threshold and the specific nature of primordial non-Gaussianities.