Gold Nanoparticle Size Estimator
Estimate AuNP size and concentration from UV-Vis spectra using the Haiss method
Sizing Method
⚡ Auto-UpdateThe absorbance ratio method is generally more robust as it doesn't require precise wavelength calibration.
UV-Vis Spectral Data
Wavelength at the maximum of the surface plasmon resonance peak.
Maximum absorbance value at the SPR peak wavelength.
Absorbance at the interband transition region. Used for both sizing and concentration.
Standard cuvettes are 1 cm. Microvolume instruments may use shorter paths.
Data quality: Use blank-subtracted absorbance values. Ensure the sample is well-dispersed (no aggregation) and the absorbance is within the linear range of your instrument (typically 0.1–2 AU).
Results
Validation: Results should be verified with TEM or DLS for critical applications. Polydispersity and non-spherical particles can lead to inaccurate estimates.
The Haiss Method for AuNP Characterization
The Haiss method, published in Analytical Chemistry in 2007, provides a rapid, non-destructive approach to estimate gold nanoparticle size and concentration directly from UV-Vis spectra. It exploits the size-dependent optical properties arising from the surface plasmon resonance (SPR) of gold nanoparticles.
Key insight: As gold nanoparticle size increases, the SPR peak redshifts (moves to longer wavelengths) and the ratio ASPR/A450 increases due to enhanced scattering contributions at larger sizes.
Sizing Methods
Method 1: SPR Peak Position (35-100 nm)
For particles larger than 35 nm, the SPR peak position provides reliable sizing. The Haiss forward equation relates size to wavelength:
λSPR = λ₀ + L₁ × exp(L₂ × d)
Inverted to solve for diameter:
d = ln((λSPR − λ₀) / L₁) / L₂
where L₁ = 6.53, L₂ = 0.0216, λ₀ = 512 nm
This method achieves ~3% error for spherical particles in water but is sensitive to refractive index changes from surface coatings or different solvents.
Mathematical constraint: The SPR peak method requires λSPR > 518.5 nm. Below this threshold, the logarithm yields undefined or negative values. For small particles with λSPR in the 510-518 nm range, use the Absorbance Ratio method instead.
Method 2: Absorbance Ratio (5-80 nm)
The ratio of SPR absorbance to the interband transition absorption at 450 nm provides a more robust sizing method:
d = exp(B₁ × (ASPR/A450) − B₂)
where B₁ = 3.00, B₂ = 2.20
This method is less sensitive to concentration errors and achieves ~6% accuracy. It works well for smaller particles where the SPR peak position is less size-dependent.
Concentration Determination
Particle concentration is calculated from the absorbance at 450 nm, which is relatively independent of particle shape and aggregation:
N = A450 × 10¹⁴ / d² (particles/mL)
This can be converted to molar concentration by dividing by Avogadro's number (6.022 × 10²³).
Why 450 nm?
The wavelength 450 nm falls in the interband transition region of gold, where:
- Absorption is dominated by electronic transitions rather than plasmon resonance
- The extinction coefficient scales predictably with particle volume
- The signal is insensitive to particle shape, aggregation state, and surface chemistry
Limitations
- Spherical particles only: Rods, cubes, and other shapes have fundamentally different optical properties
- Aqueous dispersions: Different solvents shift λSPR significantly due to refractive index changes
- Monodisperse samples: Polydispersity leads to peak broadening and inaccurate size estimates (biased toward larger particles)
- No aggregation: Aggregated samples show red-shifted, broadened peaks that invalidate the analysis
- Size limits: Below ~5 nm, the SPR is severely damped; above ~100 nm, multipolar modes complicate the spectrum
Surface coating effects: Thick coatings (>2 nm) with high refractive index can redshift λSPR by 5-10 nm, leading to size overestimation. For functionalized particles, the absorbance ratio method is preferred.
References
AI & Computational Tools for Researchers
Explore our curated guides to the best free AI tools for literature discovery, data analysis, computational modeling, and more.
