Reference ID: MET-369E | Process Engineering Reference Sheets Calculation Guide
Introduction & Context
The Residence Time Distribution (RTD) characterises how long different fluid elements remain inside a continuous-flow vessel. The exit-age density function, E(t), is the cornerstone of RTD analysis: its integral over any time interval gives the fraction of fluid that has spent between t and t+dt inside the system. Knowledge of E(t) allows process engineers to diagnose non-idealities (bypassing, dead zones), size reactors, predict conversion, and scale-up from pilot to industrial units.
Methodology & Formulas
Geometric and flow variables
Pipe diameter: D [m] (converted from mm)
Volumetric flow rate: Q [m3 s-1] (converted from L min-1)
Cross-sectional area: A = \(\dfrac{\pi D^{2}}{4}\)
Mean velocity: u = \(\dfrac{Q}{A}\)
Reynolds number: Re = \(\dfrac{\rho u D}{\mu}\)
Reactor volume: V = A L
Nominal space time: τnom = \(\dfrac{V}{Q}\)
Tracer response and normalisation
Raw concentration data: Craw(t) [mg L-1] (sampled every Δt)
Total injected mass recovered: \(\displaystyle\text{Area}_{C} = \int_{0}^{\infty}C_{\text{raw}}(t)\,dt \approx \sum_{i}\tfrac{1}{2}\left[C_{\text{raw},i} + C_{\text{raw},i-1}\right]\Delta t\)
Exit-age density: E(t) = \(\dfrac{C_{\text{raw}}(t)}{\text{Area}_{C}}\) [s-1]
Integral properties
Normalisation check: \(\displaystyle\int_{0}^{\infty}E(t)\,dt \approx \sum_{i}\tfrac{1}{2}\left[E_{i} + E_{i-1}\right]\Delta t = 1\)
Mean residence time: \(\displaystyle t_{\text{mean}} = \int_{0}^{\infty}t\,E(t)\,dt \approx \sum_{i}t_{i}\,E_{i}\,\Delta t\)
Cumulative distribution: \(\displaystyle F(t) = \int_{0}^{t}E(t')\,dt'\)
Validity regimes & acceptance criteria
Parameter
Condition
Threshold
Consequence if violated
Reynolds number
Laminar assumption
Re < 2000
Axial dispersion model required; provided E(t) shape invalid
Normalisation
Mass conservation
\(\left|\int E(t)\,dt - 1\right| \le 0.01\)
Tracer mass balance error > 1%; re-check data or numerical integration
Systematic deviation > 5%; possible calibration or bypass issue
The E(t) function, or Residence Time Distribution, tells you how long individual fluid elements actually stay inside your vessel. Unlike ideal plug-flow or CSTR assumptions, real equipment has a spectrum of residence times. Knowing E(t) lets you predict conversion, selectivity, and even product quality in reactors, crystallizers, and dryers.
Inject a non-reactive tracer pulse or step at the inlet.
Measure tracer concentration versus time at the outlet using an inline analyzer or grab samples.
Normalize the concentration curve to obtain E(t) = C(t)/∫C(t)dt for pulse, or E(t) = dF(t)/dt for step.
Use a tracer compatible with your process—NaCl for water, helium for gases, or radioisotopes for high-temperature units.
A long tail indicates dead zones or bypassing. For reaction orders >1, the tail lowers conversion because fresh reactant short-circuits; for orders <1, the tail can actually help because unreacted material eventually reacts. Quantify the impact by integrating the rate equation weighted by E(t): X = ∫[1-exp(-kτCA0n-1)]E(τ)dτ.
Both. For new designs, run a cold-flow pilot unit at the same Re and geometry to obtain E(t), then scale-up using the same dimensionless distribution. For existing units, tweak internals—baffles, screens, or distributor plates—to narrow E(t) and push performance closer to ideal.
Worked Example – RTD of a Small Process Line
A pilot–plant reactor is modelled as a straight 50 mm I.D. pipe, 2 m long, through which 3.9 L min–1 of water at 25 °C is pumped. A pulse of tracer is injected at the inlet and the outlet concentration is recorded every 2 s. Demonstrate that the resulting residence–time distribution is physically consistent and determine the mean residence time.
Calculate the internal cross-sectional area:
\[ A = \frac{\pi D^{2}}{4} = \frac{\pi (0.050)^{2}}{4} = 0.00196\ \text{m}^{2} \]
Determine the mean velocity:
\[ u = \frac{Q}{A} = \frac{6.5 \times 10^{-5}}{0.00196} = 0.033\ \text{m s}^{-1} \]
Compute the Reynolds number:
\[ Re = \frac{\rho u D}{\mu} = \frac{1000 \times 0.033 \times 0.050}{0.001} = 1655 \]
Since Re < 2300, flow is laminar.
Calculate the internal volume:
\[ V = A L = 0.00196 \times 2.0 = 0.00393\ \text{m}^{3} \]
Evaluate the nominal mean residence time:
\[ \tau_{\text{nom}} = \frac{V}{Q} = \frac{0.00393}{6.5 \times 10^{-5}} = 60.4\ \text{s} \]
Normalise the experimental tracer curve C(t) to obtain the RTD function E(t) and verify unit area:
\[ \int_{0}^{\infty} E(t)\,dt = \sum E(t)\Delta t = 1.00 \]
Compute the experimental mean residence time:
\[ t_{\text{mean}} = \int_{0}^{\infty} t E(t)\,dt = \sum t E(t)\Delta t = 40.4\ \text{s} \]
Final Answer
The nominal mean residence time is 60.4 s, while the first moment of the measured RTD gives tmean = 40.4 s. The dimensionless check ∑E(t)Δt = 1.00 confirms that the tracer response is correctly normalised and the data set is internally consistent.
"Un projet n'est jamais trop grand s'il est bien conçu."— André Citroën
"La difficulté attire l'homme de caractère, car c'est en l'étreignant qu'il se réalise."— Charles de Gaulle
