Home » Calcs » LNP Formulation: N/P Ratio, Lipid Volumes and Mixer Flows

LNP Formulation: N/P Ratio, Lipid Volumes and Mixer Flows

Calculate RNA phosphate moles, N/P-based lipid stock volumes, and mixer flows. Track ethanol through mixing, dilution, and ideal equilibrium dialysis.

Wired White

Published 28. October 2026 · version 1

#dialysis #dilution #herringbone mixer #lipid stocks #LNP #microfluidic mixing #mRNA #N/P ratio

This LNP formulation calculator turns an RNA batch into a transparent material and flow balance. It is intended for students, formulation scientists, and engineers planning research-scale lipid nanoparticle mixing. Start with RNA mass, or enter RNA length together with the molar amount of RNA molecules. Length by itself does not determine a batch size.

The N/P ratio calculation uses approximately one phosphate per nucleotide. With the same average residue molar mass, a fixed RNA mass contains the same phosphate amount regardless of strand length. The default 330 g/mol is an approximation for nucleotide residues in a polymer; do not use the molar mass of free nucleotide triphosphates. Caps, terminal groups, modified residues, and sequence composition can require a better mass model. Enter the number of countable ionizable amines per lipid explicitly. N/P is a formulation stoichiometry, not a measurement of the particle's charge.

Enter the four lipid components as molar parts, in this order: ionizable lipid, helper phospholipid, cholesterol, and PEG lipid. The calculator normalizes their sum. The illustrative 50:10:38.5:1.5 ratio is reported in the literature; it does not select a particular proprietary lipid. Enter stock molarities from your own preparation records. The results give each lipid amount and the corresponding stock volume, then the additional organic solvent required.

RNA concentration in the aqueous feed sets its batch volume. The aqueous-to-organic flow-rate ratio determines the organic volume. The default lipid nanoparticle mixing ratio is 3:1, giving 25 vol% ethanol immediately after mixing when the organic feed is pure ethanol. Total flow sets the two pump rates and ideal batch duration. If the stocks cannot fit into the required organic volume, reduce the aqueous RNA concentration, reduce the flow ratio, or increase the lipid stock concentrations. A separate balance calculates the RNA stock and aqueous buffer volumes.

Choose an ethanol target to calculate post-mixer buffer addition. Targets above the existing concentration simply require no dilution. The charts show the accompanying decrease in RNA concentration and how fresh dialysis contacts reduce freely diffusible ethanol in an ideal equilibrium model. Each contact includes a fresh ethanol-free bath; the first bath counts as one contact. Dialysis leaves sample volume unchanged in this estimate.

The animated herringbone schematic illustrates advection and folding of the two streams. It is qualitative and does not calculate mixing time, pressure drop, particle size, or encapsulation efficiency. Volumes assume additive solutions, and the batch balance excludes mixer dead volume, priming waste, adsorption, RNA loss, and later concentration. Measure RNA recovery, particle properties, and residual ethanol experimentally. These calculations support research planning and teaching; they are not a validated manufacturing process.

How it is calculated

RNA and N/P stoichiometry

Use the long-RNA convention of one phosphate per nucleotide residue. If mass $m_R$ is supplied, $n_P=m_R/M_{nt}$ . If molecular amount $n_R$ and length $L$ are supplied, $n_P=n_R L$ and $m_R=n_R L M_{nt}$ . All mass and amount conversions are explicit. A length alone cannot specify a batch. The default residue molar mass is the common 330 g/mol approximation, also used in the primary study doi:10.1038/s41598-023-42274-z (quantitative formalisms).

For a target ratio $R_{NP}$ and $z_N$ countable amines per ionizable lipid, $n_I=R_{NP}n_P/z_N$ . With lipid molar parts $r_I,r_H,r_C,r_G$ , each other component is $n_j=n_I r_j/r_I$ . Divide component moles by its stock molarity to obtain the stock volume. McKenzie et al. (2023), Basic Protocol 2, defines this N/P convention and reports the illustrative 50:10:38.5:1.5 ratio. Only published ratios are cited for proprietary lipids; the tool uses generic components and user-supplied stock molarities.

Mixer and dilution balances

The prepared aqueous-feed concentration gives $V_A=m_R/C_R$ . For $F=Q_A/Q_O$ , $V_O=V_A/F$ , $Q_O=Q_T/(1+F)$ and $Q_A=FQ_O$ . Organic top-up is $V_O-\sum_j V_j$ ; aqueous buffer addition is $V_A-m_R/C_{stock}$ . Negative additions are infeasible and block results.

Belliveau et al. (2012), attached PDF p. 7, uses a 3:1 aqueous-to-ethanol ratio in an SHM. Kulkarni et al. (2018), attached PDF p. 7, uses 3:1 and describes subsequent dialysis. The ethanol balance is $p_{mix}=p_E/(1+F)$ with percentage-point values. To reach a target by adding ethanol-free buffer, use $D=\max(1,p_{mix}/p_{target})$ , $V_{dil}=D(V_A+V_O)$ , and $p_{dil}=p_{mix}/D$ . RNA concentration decreases by the same factor. Maeki et al. (2017), attached PDF p. 12, independently reports 25% and 10% ethanol at flow ratios 3 and 9.

Ideal finite-bath dialysis

For fresh ethanol-free bath volume $V_B$ and unchanged sample volume $V_S$ , conservation at complete equilibrium gives $C_{next}/C_{before}=V_S/(V_S+V_B)=1/(1+B)$ , where $B=V_B/V_S$ . After $k$ fresh contacts, $p_{final}=p_{dil}(1+B)^{-k}$ . The first bath is contact one; two contacts mean one replacement. Thermo Fisher Tech Tip 20 (attached p. 1) explains repeated equilibration; its 200-fold approximation for a 200:1 bath is 201-fold in this exact finite-volume balance. No dialysis rate model is used.

Assumptions

RNA is a long polymer with approximately one backbone phosphate per nucleotide; end and cap corrections are neglected. The 330 g/mol residue average is adjustable. Mass means RNA analyte mass, excluding unrelated salts or buffer.
N/P counts the amines specified by the user, not all nitrogen atoms and not the pH-dependent number of protonated sites. Only the ionizable component contributes to N in this model.
Ratios are molar parts normalized by their sum. The illustrative four-component ratio and example stock concentrations do not prescribe a proprietary lipid or guarantee encapsulation.
Each lipid stock and the organic top-up share the entered ethanol fraction. Aqueous feed and dilution buffer contain no ethanol. Solution volumes are additive; solvent contraction is neglected.
Batch mass is the RNA actually entering the mixer. All feed is collected; dead volume, priming waste, recovery losses, adsorption, evaporation, RNA degradation, and lipid composition drift are excluded.
Dialysis assumes a freely permeating ethanol solute, complete equilibration on every contact, fresh ethanol-free baths, and constant sample volume. RNA remains fully retained. The finite-bath equation is derived here by conservation.
Animation paths and herringbone geometry are original qualitative illustrations. Their speed and dimensions are unrelated to physical residence time or CFD.

Where it applies

Use for research planning and teaching with known RNA and stock concentrations. Belliveau and Kulkarni studied siRNA; their mixing and solvent accounting informs this tool, but biological performance does not transfer automatically to mRNA. McKenzie (2023) supplies the mRNA N/P convention and illustrative ratios.

The calculation does not predict encapsulation efficiency, size distribution, internal morphology, potency, mixing time, pressure drop, buffer pH, solubility, or stability. The 3–8 mRNA N/P range is literature context rather than a universal operating limit. Select and validate your own lipids, buffers, flow rates, and process.

An organic stock-volume excess cannot be fixed by entering a negative top-up: increase stock concentrations, reduce aqueous-feed RNA concentration, or reduce FRR. Aqueous-feed RNA concentration must not exceed RNA stock concentration. Dialysis output is an equilibrium estimate, not a residual-solvent specification or a time-to-target prediction. Real volume changes, membrane transport limitations, lipid exchange, and losses require measurement. Measure final volume, RNA recovery, particle properties, and residual ethanol experimentally; this is not a validated manufacturing or clinical preparation protocol.

References

Belliveau NM et al. Microfluidic Synthesis of Highly Potent Limit-size Lipid Nanoparticles for In Vivo Delivery of siRNA. Molecular Therapy–Nucleic Acids 1, e37 (2012). https://doi.org/10.1038/mtna.2012.28 — Methods: SHM, 3:1 aqueous:ethanol mixing and dialysis.
Kulkarni JA et al. On the Formation and Morphology of Lipid Nanoparticles Containing Ionizable Cationic Lipids and siRNA. ACS Nano 12, 4787–4795 (2018). https://doi.org/10.1021/acsnano.8b01516 — attached author-hosted PDF p. 7: mixing and buffer exchange.
McKenzie RE et al. mRNA Synthesis and Encapsulation in Ionizable Lipid Nanoparticles. Current Protocols 3, e898 (2023). https://doi.org/10.1002/cpz1.898 — Basic Protocol 2: amine/phosphate definition, multiple amines, and published lipid ratios.
Maeki M et al. Understanding the formation mechanism of lipid nanoparticles in microfluidic devices with chaotic micromixers. PLOS ONE 12, e0187962 (2017). https://doi.org/10.1371/journal.pone.0187962 — attached PDF p. 12: 25% and 10% ethanol at FRR 3 and 9.
Thermo Fisher Scientific. Tech Tip 20: Dialysis—an overview (2008). https://www.thermofisher.com/TFS-Assets/LSG/brochures/TR0020-Dialysis-overview.pdf — p. 1: repeated finite-bath equilibration and approximate worked reduction factors.
Quantitative size-resolved characterization of mRNA nanoparticles by in-line coupling of asymmetrical-flow field-flow fractionation with small angle X-ray scattering. Scientific Reports (2023). https://doi.org/10.1038/s41598-023-42274-z — quantitative formalisms use 330 Da per negatively charged RNA nucleotide.

Inputs

Input	Symbol	Unit	Default
RNA input basis Length alone is not an amount: the second mode also requires picomoles of RNA molecules.	`rna_basis`		RNA mass
RNA mass	`m_R`	ug	100
RNA length (nucleotides) Include untranslated regions and the poly(A) tail. One phosphate per nucleotide is the long-RNA approximation.	`L_R`		4000
Amount of RNA molecules	`a_R`	pmol	75.75757576
Mean nucleotide residue molar mass 330 g/mol is a common approximation, not the mass of a free nucleotide triphosphate. Use a sequence-specific residue average when known.	`M_nt`	g/mol	330
Target N/P ratio Moles of countable ionizable amines divided by RNA phosphate moles; not a prediction of net charge at a particular pH.	`NP`		6
Countable amines per ionizable lipid Count the ionizable amines used in your N/P convention, not every nitrogen in the molecular formula.	`z_N`		1
Ionizable lipid — molar parts Ratios are normalized by their sum; they need not sum to 100. Published ratios are illustrative, not a validated formulation.	`r_I`		50
Helper phospholipid — molar parts	`r_H`		10
Cholesterol — molar parts	`r_C`		38.5
PEG lipid — molar parts	`r_G`		1.5
Ionizable lipid stock concentration Enter the molarity of your actual stock. All four stocks and the top-up solvent must have the same ethanol fraction.	`C_I`	mmol/L	25
Helper phospholipid stock concentration	`C_H`	mmol/L	10
Cholesterol stock concentration	`C_C`	mmol/L	25
PEG lipid stock concentration	`C_G`	mmol/L	2.5
RNA concentration in aqueous feed The concentration after aqueous-feed preparation. It determines the feed volumes for the selected RNA batch.	`C_R`	ug/mL	100
RNA stock concentration Default: 1000 µg/mL (1 mg/mL). Enter 0.1–50,000 µg/mL.	`C_stock`	ug/mL	1000
Aqueous : organic flow-rate ratio FRR = aqueous flow / organic flow. The usual 3:1 setting gives 25 vol% ethanol for a pure-ethanol organic feed.	`FRR`		3
Total mixer flow	`Q_T`	mL/min	4
Ethanol in organic feed (vol%) All lipid stocks and organic top-up solvent share this composition. Aqueous feed and dilution buffer are ethanol-free.	`p_E`		100
Ethanol target after dilution (vol%) Buffer is added only if the mixed suspension exceeds this target.	`p_target`		5
Dialysis bath / sample volume Fresh, ethanol-free bath volume for each contact divided by the diluted sample volume.	`B_D`		200
Complete dialysis contacts First bath counts as one contact; one buffer replacement gives two contacts. Each is assumed to reach equilibrium. Zero skips dialysis.	`k_D`		2
Mixer animation time Illustrative folding and advection through herringbone grooves; not CFD, residence time, or a particle-size prediction.	`t_anim`	s	2.4

Results

With the default inputs above, the calc gives:

Result	Value
RNA phosphate amount	0.303 umol
Total lipid amount	3.636 umol
Aqueous flow	3 mL/min
Organic flow	1 mL/min
Ideal batch duration	20 s
Mixed batch volume	1.333 mL
Ethanol after mixing (vol%)	25
Dilution buffer to add	5.333 mL
Sample volume after dilution	6.667 mL
RNA after dilution / ideal dialysis	15 ug/mL
Ethanol after dilution (vol%)	5
Ideal ethanol after dialysis (vol%)	1.238 × 10⁻⁴
Fresh dialysis bath per contact	1,333 mL
Total fresh dialysis buffer	2,667 mL

Checked against sources

Each case below was calculated by this calc and agrees with the value from its source (within the tolerance where one is shown).

Published ethanol at FRR 3

Source: Maeki et al. (2017), doi:10.1371/journal.pone.0187962, PDF p. 12: complete-mixing ethanol fraction at aqueous:ethanol FRR 3. Independent published solvent-balance value; not a prediction of particle size.

Inputs: the default inputs above.

Result	Source	This calc
Ethanol after mixing (vol%)	25	25

Published ethanol at FRR 9

Source: Maeki et al. (2017), doi:10.1371/journal.pone.0187962, PDF p. 12. Lower RNA feed concentration is a batch-planning choice to keep lipid stock volumes feasible; the expected 10% ethanol comes directly from the paper.

Inputs: Aqueous : organic flow-rate ratio = 9; RNA concentration in aqueous feed = 30 ug/mL; the others as above.

Result	Source	This calc
Ethanol after mixing (vol%)	10	10

Published three-contact dialysis approximation

Source: Thermo Fisher Tech Tip 20, p. 1, 1 mL sample with 200 mL fresh bath on three equilibrations: approximately 8,000,000-fold reduction. The source neglects the 1 mL sample in total volume. Exact conservation gives 201³ = 8,120,601; tolerance 130,000 (1.625%) explicitly allows that approximation. Equations retain the sample volume.

Inputs: Complete dialysis contacts = 3; the others as above.

Result	Source	This calc
Ideal dialysis concentration reduction factor	8,000,000 ± 130,000	8,120,601

Published 1.125 mL aqueous feed for 1.5 mL batch

Source: McKenzie et al. (2023), doi:10.1002/cpz1.898, Alternate Protocol step 5 explicitly states that 1.5 mL total formulation requires at least 1.125 mL aqueous syringe volume at FRR 3:1. This check tests only the published hydraulic split; a 112.5 µg RNA batch at 100 µg/mL is a chosen encoding of that aqueous volume. It does not reproduce the paper's preformed-vesicle biology.

Inputs: RNA mass = 112.5 ug; Total mixer flow = 12 mL/min; the others as above.

Result	Source	This calc
Aqueous feed volume	1.125 mL	1.125 mL
Mixed batch volume	1.5 mL	1.5 mL

LNP Formulation: N/P Ratio, Lipid Volumes and Mixer Flows