CoChat

A Tutorial on THz Pulse Analysis: Accurate Retrieval of Pulse Arrival Times, Spectral Energy Density and Absolute Spectral Phase

Open access
Lloyd-Hughes, James, Chopra, Nishtha, Deveikis, Justas, Pandya, Raj, Woolley, Jack · 2025 · Journal of Infrared, Millimeter, and Terahertz Waves

The DOI matches its Crossref record; 44 of the 44 references it cites were found in a registry or an index; 14 of the 34 sources its claims cite were checked for support.

Abstract

Electro-optic sampling allows the electric field of THz, mid-infrared and visible light pulses to be measured directly as a function of time, with data analysis often performed in the frequency domain after fast Fourier transform. Here, we review aspects of Fourier theory relevant to the frequency-domain analysis of light pulses recorded in the time domain. We describe a “best practise” approach to using the discrete Fourier transform that ensures consistency with analytical results from the continuous Fourier transform. We summarise a phenomenological time-domain model of THz pulses, based on carrier and envelope waves, and show that it can reproduce a wide variety of experimental single- to multi-cycle THz pulses, with exemplary data from lab-based sources (photoconductive antennae, optical rectification, spintronic emitters) and a THz free-electron laser. A quantitative comparison of the spectral energy density of these distinct sources is enabled by the amplitude-accurate discrete Fourier transform. We describe a method that ensures the accurate calculation of the absolute spectral phase (valid for arbitrary sampling windows in the time domain) and summarise how the carrier-envelope phase, pulse arrival time, and chirp can be obtained from the phase. Our aim with this overview of THz pulse analysis is to highlight algorithms and concepts that are useful to newcomers to time-domain spectroscopy and experts, alike.

Overview

CoChat Summary

TL;DR

This tutorial reviews Fourier-transform best practices for time-domain THz pulse analysis, providing an amplitude- and phase-accurate discrete Fourier transform and a phenomenological carrier-envelope pulse model. It demonstrates how accurate absolute spectral phase and amplitude scaling enable retrieval of pulse arrival times, carrier-envelope phase, chirp, and quantitative spectral energy density comparisons across THz sources.

Key findings

  • A phase-corrected DFT, \(\tilde{E}(\omega_k)=\delta t \sum_m E_m e^{i\omega_k t_m} e^{-i\omega_k T_0}\), reproduces the continuous Fourier transform’s amplitude and absolute phase, whereas the standard DFT returns a spurious linear phase that depends on the sampling window; this was validated for a Gaussian pulse with \(\sigma=200\) fs and \(T_0=1\) ps.
  • A carrier-envelope model with a sine carrier and sech or Gaussian envelope reproduces single- and multi-cycle THz pulses; for a sech envelope the high-frequency amplitude spectrum decays exponentially with slope \(d|\tilde{E}|/d\omega=-\pi\tau/2\), allowing the envelope duration \(\tau\) to be inferred.
  • Spectral energy density comparisons: TELBE free-electron laser at 0.68 THz, 50 kHz, \(\sim32\) mW, \(E_{\text{pulse}}=0.65\ \mu\)J, peak field \(\sim87\) kV/cm, max \(u(f)=13.5\ \text{J m}^{-2}\text{THz}^{-1}\); LiNbO\(_3\) optical rectification gave \(\sim650\) kV/cm, \(E_{\text{pulse}}=1\ \mu\)J, \(u(f)=2.5\ \text{J m}^{-2}\text{THz}^{-1}\); a spintronic emitter gave \(\sim50\) kV/cm, \(E_{\text{pulse}}=3.8\) nJ, with lower energy density below 3.8 THz but broader bandwidth.
  • For an experimental photoconductive emitter, the absolute phase started near \(-\pi/2\), and after detector-response deconvolution the phase was flatter; a time-domain fit yielded \(t_e=10.1\pm0.2\) fs, \(\tau=83\pm1\) fs, \(f_c=1.89\pm0.01\) THz, and \(t_{\text{CEP}}=-35\pm1\) fs.

Methods

  • Tutorial/review design: derives amplitude- and phase-correct forward and inverse DFT expressions, validates them against analytical continuous Fourier transform results for Gaussian and sech pulses, and provides Python implementation in an appendix.
  • Phenomenological time-domain model: electric field as carrier wave \(-\sin(\omega_c(t-t_e)-\phi_0)\) times a Gaussian or sech envelope, extended for finite carrier-envelope phase using an analytic representation with Heaviside step function and global phase factor.
  • Experimental data: THz-TDS waveforms from a photoconductive antenna, optical rectification in LiNbO\(_3\), a spintronic emitter, and the TELBE free-electron laser; electro-optic sampling with GaP; spectral energy density calculated via Parseval’s theorem; literature reports of spectral phase also reviewed.

Limitations

  • The paper states the carrier-envelope model is not based on generation physics or propagation and does not produce realistic pulses for short envelopes (\(\tau \le 2\pi/\omega_c\)) when the carrier-envelope phase makes the carrier close to a cosine.
  • The paper states the model fit did not capture the more complex TELBE pulse structure at times above 5 ps, and that arrival-time methods have limitations: the educated-guess method fails for finite CEP, the mean-arrival-time method is pulled by pulse echoes, and the group-delay method stumbles if the spectral phase is complex near maximum amplitude.
  • Evident from the text: the literature survey of spectral phase is described as representative but non-exhaustive, and the experimental comparison is limited to selected sources; detector-response correction was only possible below 7.5 THz because the electro-optic signal dropped below the noise floor.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1007/s10762-025-01052-w

Cites

44 of 44 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Image 1
From the PDF
Fig. 2 a Single-cycle THz pulse (Eq.5 with A = 1, t e = t CEP = 0, τ =100fs) sampled from -2 to 2 ps with N = 200 points (blue), and the same pulse sampled from -1.5 to 2.5 ps (orange, offset vertically for clarity). b Time series E a , b m for the data in a , as a function of index m . c Time series after applying the ifftshift algorithm. d Phase of DFT of time-shifted E a m for N = 200 (blue crosses), N = 201 (black dashed line) and E b m (orange points, N = 200). The symmetric time window deviates from the expected ± π/ 2 by a small linear term. The asymmetric time window has a larger linear slope as the ifftshift algorithm did not put the pulse centre at index m = 0. e Phase retrieved using Eq.2 acting on the unshifted data in panel b , illustrating accurate phase retrieval
Fig. 2 a Single-cycle THz pulse (Eq.5 with A = 1, t e = t CEP = 0, τ =100fs) sampled from -2 to 2 ps with N = 200 points (blue), and the same pulse sampled from -1.5 to 2.5 ps (orange, offset vertically for clarity). b Time series E a , b m for the data in a , as a function of index m . c Time series after applying the ifftshift algorithm. d Phase of DFT of time-shifted E a m for N = 200 (blue crosses), N = 201 (black dashed line) and E b m (orange points, N = 200). The symmetric time window deviates from the expected ± π/ 2 by a small linear term. The asymmetric time window has a larger linear slope as the ifftshift algorithm did not put the pulse centre at index m = 0. e Phase retrieved using Eq.2 acting on the unshifted data in panel b , illustrating accurate phase retrievalFrom the PDF
Fig. 3 THzpulses calculated for sech (blue lines) or Gaussian (orange lines) pulse envelopes. a Indicates the time-domain waveforms (solid lines) for a sech envelope (blue dashed line) or Gaussian envelope (orange dashed line). A 1THz carrier frequency was assumed, with zero carrier and envelope offset times, and τ = σ = 100fs. b Amplitude spectra from the DFT of the THz pulses in a , along with the analytical result for the CFT of a sech pulse (dashed line). c Phase spectra of data in a from DFT and CFT
Fig. 3 THzpulses calculated for sech (blue lines) or Gaussian (orange lines) pulse envelopes. a Indicates the time-domain waveforms (solid lines) for a sech envelope (blue dashed line) or Gaussian envelope (orange dashed line). A 1THz carrier frequency was assumed, with zero carrier and envelope offset times, and τ = σ = 100fs. b Amplitude spectra from the DFT of the THz pulses in a , along with the analytical result for the CFT of a sech pulse (dashed line). c Phase spectra of data in a from DFT and CFTFrom the PDF
Fig. 4 Impact of pulse duration. a Time-domain THz pulses for a sech envelope with different pulse durations, τ = 50fs to 2000fs (coloured lines). Waveforms are offset for clarity. Dashed lines indicate the carrier wave (1THz frequency, or 1ps period). b Amplitude spectra of the THz pulses in a
Fig. 4 Impact of pulse duration. a Time-domain THz pulses for a sech envelope with different pulse durations, τ = 50fs to 2000fs (coloured lines). Waveforms are offset for clarity. Dashed lines indicate the carrier wave (1THz frequency, or 1ps period). b Amplitude spectra of the THz pulses in aFrom the PDF
Fig. 5 Influence of pulse arrival time and carrier-envelope phase. a THz time-domain pulses calculated using Eq. 5 with ω/ 2 π = 1THz, τ = 100fs, t CEP = 0 and various values of t e . Black points indicate the pulse arrival time (mean arrival time method). Pulses are plotted offset vertically for clarity. b Unwrapped phase spectra from DFT of data in a . c THz pulses with t e = 0, and varying carrier times 0 ≤ t c ≤ 500fs, corresponding to different carrier-envelope phases 0 ≤ φ 0 ≤ π . The dashed lines indicate the sech envelope and the black points are the mean arrival time. d Phase spectra corresponding to the cases shown in c
Fig. 5 Influence of pulse arrival time and carrier-envelope phase. a THz time-domain pulses calculated using Eq. 5 with ω/ 2 π = 1THz, τ = 100fs, t CEP = 0 and various values of t e . Black points indicate the pulse arrival time (mean arrival time method). Pulses are plotted offset vertically for clarity. b Unwrapped phase spectra from DFT of data in a . c THz pulses with t e = 0, and varying carrier times 0 ≤ t c ≤ 500fs, corresponding to different carrier-envelope phases 0 ≤ φ 0 ≤ π . The dashed lines indicate the sech envelope and the black points are the mean arrival time. d Phase spectra corresponding to the cases shown in cFrom the PDF
Image 6
From the PDF
Fig. 6 a Multi-cycle THz pulse produced by TELBE, running at 0.68THz. The black spot is the pulse arrival time calculated from the mean-time method. b Electric field produced by LiNbO 3 photoexcited by a 1mJ, 800nm, 40fs laser pulse. c As b but for a spintronic emitter. d Spectral energy density u ( f ) for the THz sources in a -c , shown over frequency range where the signal exceeded the noise floor (which varied from experiment to experiment)
Fig. 6 a Multi-cycle THz pulse produced by TELBE, running at 0.68THz. The black spot is the pulse arrival time calculated from the mean-time method. b Electric field produced by LiNbO 3 photoexcited by a 1mJ, 800nm, 40fs laser pulse. c As b but for a spintronic emitter. d Spectral energy density u ( f ) for the THz sources in a -c , shown over frequency range where the signal exceeded the noise floor (which varied from experiment to experiment)From the PDF
Image 8
From the PDF

Verification

Claims the paper makes

10 Valid · 1 Not valid · 11 Not checked · 74 Own results and interpretations

36 more claims

Create a free account to see all 96 claims and to verify the 4 references still open.

See all claims — free account
  • Probably not a substantial claim. 20
  • It may state two separate claims as one. 45
  • The quote only partly states this claim. 1

Create a free account to see these claims.

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

2 of 2

Citations

Cites

Show all 44 references

Cited by

See all 6

Metadata

Authors
Lloyd-Hughes, J., Chopra, N., Deveikis, J., Pandya, R., & Woolley, J.
Affiliations
University of Warwick (GB)
Venue
Journal of Infrared, Millimeter, and Terahertz Waves
Published
2025-05-16
Volume
46
Issue
6
ISSN
  • 1866-6892 (ISSN-L)
  • 1866-6906
OpenAlex
W4410452786
Cited by
6
References
38
Citations by year
2026: 5 · 2025: 1
FWCI
1.35
Type
journal article
Language
en
Open access
Licence
Locations
  • Journal of Infrared Millimeter and Terahertz Waves (publishedVersion, cc-by)
  • arXiv (Cornell University) (submittedVersion)
  • Figshare (submittedVersion, public-domain)
  • arXiv (Cornell University)
Funders
  • EPSRC (EP/V047914/1)
  • Horizon Europe (101103873)
Topics
  • Physical Sciences › Engineering › Electrical and Electronic Engineering › Terahertz technology and applications
  • Physical Sciences › Physics and Astronomy › Atomic and Molecular Physics, and Optics › Laser-Matter Interactions and Applications
  • Physical Sciences › Engineering › Electrical and Electronic Engineering › Particle Accelerators and Free-Electron Lasers
Keywords
electro-optic sampling, discrete Fourier transform, continuous Fourier transform, spectral energy density, carrier-envelope phase, pulse arrival time, chirp, photoconductive antenna, optical rectification, spintronic emitters, time-domain spectroscopy

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol and qwen3.8-omni-flash).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: A Tutorial on THz Pulse Analysis: Accurate Retrieval of Pulse Arrival Times, Spectral Energy Density and Absolute Spectral Phase. https://app.cochat.ai/papers/10.1007/s10762-025-01052-w
The paper
Lloyd-Hughes, J., Chopra, N., Deveikis, J., Pandya, R., & Woolley, J. (2025). A Tutorial on THz Pulse Analysis: Accurate Retrieval of Pulse Arrival Times, Spectral Energy Density and Absolute Spectral Phase. Journal of Infrared, Millimeter, and Terahertz Waves. https://doi.org/10.1007/s10762-025-01052-w

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.