HILIC instead of reversed phase: when to switch

Use HILIC instead of reversed phase when polar or ionized analytes stay weakly retained after pH adjustment and aqueous-stable reversed-phase columns have been tried, and when sample solubility, diluent, detection and method intent allow a mode change.^1,2 Very low k is a warning sign, not a universal switch threshold.

Why does reversed-phase retention collapse for polar and ionized analytes?

Reversed-phase LC is particularly suited to hydrophobic compounds but fails to retain polar or charged compounds.³ Low retention of hydrophilic compounds, particularly acids and bases, has historically been one of its weaknesses.⁴

Ionization is the first thing to check. On a C18 column with 10% acetonitrile and 90% phosphate buffer at pH 2.4, the retention factor of benzoic acid (k) is about 35.⁴ At pH 6 the acid is deprotonated and k falls to around 1, a 35-fold decrease.⁴ Benzoic acid still keeps enough retention to be separated because its phenyl group adds lipophilicity; for analytes whose remaining structure is hydrophilic, adequate retention can be hard to obtain.⁴

A common warning sign in a low-pH screen is that some compounds behave normally while others sit at the column dead time; those unretained compounds are likely ionized bases.² As a working rule, Dolan adjusts k to the 2–10 range during method development.² That range is a development target, not a threshold for abandoning reversed phase. For validated or compendial methods, the applicable procedure and regulatory framework take precedence over the general rules of thumb given here.

Build it faster

AI Method Builder

Draft a starting method for your separation and refine it — save your work to your account.

Build my starting method

No account needed

How does HILIC retain compounds that reversed phase cannot?

In HILIC, solutes partition between a water layer held on the surface of a polar column and an organic-rich bulk mobile phase.^5,1 Ionic retention and adsorption can be superimposed on this partition, and the review by Buszewski and Noga describes the overall mechanism as most probably a superposition of partitioning with electrostatic interactions or hydrogen bonding.^5,1

Retention increases with solute polarity, broadly opposite to reversed phase.⁵ Water is the strong solvent: reducing its concentration increases retention, and acetonitrile is usually used at ≈60–95%.⁵ Uracil, used as a void volume marker in reversed phase, can be well retained in HILIC.⁵ The selectivity of the two modes differs considerably, which makes them complementary.⁵ Within HILIC, changing the stationary phase is probably the most effective way to change stationary-phase selectivity.⁵

Figure 1 shows the two retention models side by side.

Two particle tiles comparing reversed-phase and HILIC retention: a hydrophobic ligand retaining nonpolar analytes longest and a water-rich layer retaining polar analytes longest, with retention-order chips beneath each tile. — **Figure 1. Reversed-phase and HILIC retention models.** Reversed-phase tile: a particle with a hydrophobic bonded ligand (C18 shown), suited to hydrophobic compounds.³ HILIC tile: a polar particle holding a water-rich layer, into which polar solutes partition from the acetonitrile-rich mobile phase, with ionic retention and adsorption superimposed.^5,1 Retention-order chips show the elution order broadly opposite between the two modes; uracil, a void volume marker in reversed phase, can be well retained in HILIC.⁵ The HILIC sketch leaves the polar surface chemistry generic. Schematic sketches, not data; actual order depends on phase, pH and buffer.

When should you use HILIC instead of reversed phase?

HILIC is the mode of choice for uncharged highly hydrophilic and amphiphilic compounds that are too polar to be well retained in reversed phase.¹ Polar and ionized hydrophilic solutes are likely to be retained; as a guide, solutes with log D above ≈+1 typically show low HILIC retention, although ionic interactions can still retain them.⁵ Figure 2 groups the factors that decide the choice, and Table 1 sets out the guidance for each.

Five icon tiles for choosing reversed-phase versus HILIC: analyte polarity, ionization, sample diluent water content, MS detection and column equilibration, each with a short guidance line; the polarity tile shows a schematic log D scale with a ≈+1 guide, and two takeaways cover trying reversed-phase options first and method intent. — **Figure 2. Factors that decide between reversed phase and HILIC.** Five tiles carry the one-line guidance from Table 1: polarity and ionization (polar and ionized hydrophilic solutes are likely retained),⁵ diluent (no more than 25% water as a working rule),⁶ MS (≈3× average sensitivity gain under one set of flow-injection conditions)⁵ and equilibration (repeatable partial re-equilibration is possible; full re-equilibration can be slow).⁷ The log D scale is schematic, with only the ≈+1 guide taken from the text; the first takeaway reflects Table 1’s retention-and-selectivity row. The second takeaway states method intent: for a compendial procedure, a mode change requires revalidation. Schematic, not data; heuristics are labeled as such.

**Table 1.** Decision factors for reversed phase versus HILIC. Numeric values are working heuristics or condition-specific findings, not universal limits.^6,7
Factor	What to check	Source-based guidance
Polarity and ionization	Hydrophilicity and charge state at the mobile-phase pH	Polar and ionized hydrophilic solutes are likely retained; log D above ≈+1 typically gives low HILIC retention⁵
Retention and selectivity	Whether pH or column changes within reversed phase already give adequate retention	HILIC selectivity differs considerably from reversed phase, so the modes are complementary⁵
Sample solvent	Water content of the diluent; solubility in high organic	Diluent–eluent strength mismatch gives broad tailing or split peaks; working rule ≤25% water by volume; some compounds have low solubility in high organic^6,3
MS detection	ESI-MS required; ion-pair reagent in the current method	≈3× average sensitivity gain under one set of flow-injection conditions; alkylsulfonates contaminate ESI sources^5,4
Equilibration	Gradient cycle time and run-to-run repeatability	50–100 column volumes for a new column and 10–20 between gradient injections in one author’s experience; about two column volumes gave repeatable runs when longer than t_flush^6,7
Method intent	Compendial or validated procedure	No change of substituent identity within allowable adjustments; other changes require revalidation⁸

HILIC separations are particularly susceptible to differences between the eluotropic strength of the sample diluent and the eluent. ⁶ A mismatch in eluotropic strength between diluent and eluent can produce broad tailing or split peaks, and Taylor’s practical rule is a diluent with no more than 25% water by volume.⁶ He also advises keeping the needle-wash solvent as high in organic as possible.⁶ Low solubility of some compounds in high organic content is a recognized HILIC disadvantage.³

For ESI-MS, McCalley reported an average ≈3× sensitivity increase for 90% acetonitrile–10% 5 mM ammonium formate pH 3.0 compared with 10% acetonitrile–90% of the same buffer, measured by flow injection on a triple quadrupole for six basic and four acidic solutes.⁵ That figure belongs to those conditions, not to every analyte.

Equilibration tends to be the downfall of many first HILIC attempts.⁶ Taylor’s experience is 50–100 column volumes for a new column and 10–20 column volumes between gradient injections.⁶ Stoll and Seidl distinguish full from partial, repeatable re-equilibration: full re-equilibration of HILIC phases can take a very long time, yet 40 consecutive gradient runs were highly repeatable with about two column volumes on a conventional instrument, provided the re-equilibration time exceeded t_flush.⁷

Method intent can settle the question outright. Under the USP General Chapter ⟨621⟩ allowable adjustments, the identity of the stationary-phase substituent may not change, and changes other than those indicated require revalidation.⁸ If the environment does not allow column chemistry or separation mode to change, you are probably stuck with the ion-pairing reagents for the life of the method.⁴ For validated or compendial methods, the applicable procedure and regulatory framework take precedence over the general rules of thumb given here.

Which reversed-phase options should you try before HILIC?

Before developing an ion-pair method, Dolan advises considering mixed-mode, AQ, embedded-polar or other specialty columns.² Table 2 compares these options with HILIC.

**Table 2.** Options for weakly retained polar or ionized analytes. Qualitative comparison; the relative retention gain of AQ versus polar-embedded columns is not quantified here.²
Option	What it adds	Main drawback
Aqueous-stable (AQ) or polar-embedded column	Usable in 100% aqueous mobile phase; conventional C18 is generally kept below ≈95% water^2,4	Some bases still lack retention even in 100% aqueous mobile phase²
Long-chain ion pairing	Alkylsulfonate reagents retain positively charged bases²	20–50 column volumes or more to equilibrate; never fully washed out; alkylsulfonates contaminate ESI sources^2,4
Reversed-phase–ion-exchange mixed mode	Retains both hydrophobic and ionic compounds; lower salt than ion exchange, so MS compatible³	Often only one mechanism operates at a time, depending on the organic–aqueous ratio³
HILIC	Retains compounds too polar for reversed phase; no ion-pair reagent; ESI-MS coupling¹	Solubility in high organic, long equilibration and injection-solvent sensitivity³

AQ or embedded-polar columns under fully aqueous conditions, with buffer or pH adjustment, sometimes give enough retention for polar compounds without ion pairing.² Ion pairing also makes column history matter: Dolan dedicates a column to ion-pair work once it has been used that way.²

How do you trial and verify a HILIC method?

Figure 3 and Table 3 give the same four-step sequence, from the first sign of weak retention to verification.

Four-step workflow for switching from reversed-phase HPLC to HILIC: weak retention and ionization check, aqueous-stable column options, HILIC trial with controlled diluent, and verification of re-equilibration time and system suitability. — **Figure 3. From weak reversed-phase retention to a verified HILIC method.** Four sequential steps joined by arrows, matching Table 3 row for row: weak retention (check ionization at the mobile-phase pH),⁴ aqueous-stable options (AQ or polar-embedded column, or mixed mode),² HILIC trial (diol screening gradient, diluent ≤25% water)⁶ and verification (vary re-equilibration time, meet system suitability).^7,8 Each step carries a sketch: a dead-time versus retained peak, the 5–40% aqueous screening ramp with its three pH values, and traces showing unchanged retention at two re-equilibration times. Schematic workflow, not data; screening conditions are one practitioner’s starting point.

**Table 3.** Four-step sequence from weak reversed-phase retention to a verified HILIC method, matching Figure 3 row for row. Screening conditions are one practitioner’s starting point.⁶
Step	Action	Move on when
1. Weak reversed-phase retention	Check whether the analyte is ionized at the mobile-phase pH; low-pH peaks at dead time suggest ionized bases^4,2	pH adjustment does not give adequate retention
2. Aqueous-stable options	Try an AQ or polar-embedded column in 100% aqueous mobile phase, or a mixed-mode column²	Retention or MS compatibility is still inadequate
3. HILIC trial	Run a 5–40% aqueous gradient on a diol phase with 10 mM buffer at pH 2.5, 4 and 6; diluent ≤25% water⁶	Retention is obtained; with many critical pairs, change to an amide or bare silica phase⁶
4. Verification	Vary re-equilibration time systematically; meet system suitability throughout^7,8	Retention and resolution are stable at the chosen re-equilibration time; compendial procedures are revalidated⁸

Taylor starts with a diol phase because it cannot be ionized in the usual pH range, so the mechanism is easier to interpret, and treats ionic strength from 5 to 20 mM as a useful second variable.⁶ It is very important to maintain constant ionic strength when running gradients; both mobile phases could be prepared by adding 5% of concentrated aqueous buffer to either water or acetonitrile.⁶

Frequently asked questions

Does a k below 1 mean reversed phase has failed?

Not on its own. Benzoic acid on C18 drops from k ≈35 at pH 2.4 to around 1 at pH 6 because it ionizes, so the first check is the analyte’s charge state at the mobile-phase pH.⁴ Dolan’s 2–10 range for k is a working method-development target, not a switch threshold.² Move to aqueous-stable reversed-phase columns, mixed mode or HILIC only when pH control cannot deliver the retention and selectivity the method needs.

Why do HILIC peaks split when the sample is dissolved in water?

In HILIC, water is the strong solvent, so a water-rich diluent elutes more strongly than the starting eluent.⁵ A mismatch in eluotropic strength between diluent and eluent can give broad tailing or split peaks.⁶ As a working rule, Taylor keeps the diluent at no more than 25% water by volume and the needle wash as high in organic as possible.⁶ Altering diluent pH or ionic strength can help where solubility is the problem.⁶

Is HILIC always more sensitive than reversed phase in ESI-MS?

Not necessarily. McCalley reported an average ≈3× sensitivity gain for six basic and four acidic solutes by flow-injection ESI-MS on a triple quadrupole, comparing 90% with 10% acetonitrile in 5 mM ammonium formate pH 3.0.⁵ The gain is specific to those conditions. High-acetonitrile mobile phases favor ESI because of their low viscosity, high volatility and low surface tension, but each analyte’s response belongs to its own final conditions.⁵

How long does a HILIC column take to equilibrate?

It depends on the level of equilibration needed. In Taylor’s experience a new HILIC column needs 50–100 column volumes and gradient re-equilibration needs 10–20 column volumes.⁶ Stoll and Seidl found that full re-equilibration can take a very long time, yet repeatable gradient HILIC separations were obtained with about two column volumes on a conventional instrument when the re-equilibration time exceeded the system flush-out time.⁷ They suggest varying the re-equilibration time during development to see whether retention and resolution depend on it.⁷

Can a USP reversed-phase method be converted to HILIC as an allowable adjustment?

No. USP General Chapter ⟨621⟩ allows no change of the identity of the stationary-phase substituent, citing the replacement of C18 by C8 as an example, and changes other than those indicated require revalidation.⁸ Replacing a reversed-phase column with a HILIC phase lies well outside that scope. Whatever procedure is finally used, compliance with its system suitability criteria is required throughout the chromatographic procedure.⁸

The takeaway

Treat very low reversed-phase retention as a warning sign that calls for diagnosis, not as a trigger with a fixed k value. Check ionization first, then try aqueous-stable, polar-embedded or mixed-mode columns, and move to HILIC when the analytes are genuinely too polar for reversed phase and the sample diluent, detector and method intent allow it. A HILIC method then earns its place only after the diluent is controlled, re-equilibration time has been tested and system suitability is met; for compendial procedures, a mode change is a new procedure that requires revalidation.

References

Buszewski B, Noga S. Hydrophilic interaction liquid chromatography (HILIC)—a powerful separation technique. Analytical and Bioanalytical Chemistry. 2012;402:231–247. https://doi.org/10.1007/s00216-011-5308-5 — HILIC as the mode of choice for compounds too polar for reversed phase; partition into a water-enriched layer plus electrostatic and hydrogen-bonding contributions; pH and analyte charge state; elution order roughly reverse of reversed phase; no ion-pair reagents and ESI-MS coupling.
Dolan JW. Ion pairing — blessing or curse? LCGC North America. 2008;26(2). https://www.chromatographyonline.com/view/ion-pairing-blessing-or-curse-0 — working retention target k 2–10 (method-development heuristic); unretained ionized bases as the warning sign; AQ/polar-embedded columns in 100% aqueous; ion-pair equilibration and column-dedication drawbacks; alternatives (mixed-mode, AQ, embedded polar).
Lemasson E, Bertin S, Hennig P, Lesellier E, West C. Mixed-mode chromatography—a review. LCGC Supplements. 2017;30(6). https://www.chromatographyonline.com/view/mixed-mode-chromatography-review — reversed-phase limits for polar/charged compounds; ion-pairing MS caveat; HILIC drawbacks (solubility, equilibration, injection solvent); reversed-phase/ion-exchange mixed-mode retention and MS compatibility; mixed-mode caveat.
Stoll DR. Method from Mars? Coping with chromatographic legacies. LCGC North America. 2019;37(5). https://www.chromatographyonline.com/view/methods-mars-coping-chromatographic-legacies — ionization collapses reversed-phase retention (benzoic acid k ≈35 to ≈1); long-chain ion-pair reagents and MS; aqueous-compatible (AQ) columns; constraint of regulated methods.
McCalley DV. Hydrophilic interaction liquid chromatography: an update. LCGC Europe. 2019;32(3). https://www.chromatographyonline.com/view/hydrophilic-interaction-liquid-chromatography-update — HILIC partition mechanism and superimposed ionic retention/adsorption; retention order broadly opposite to reversed phase; which solutes suit HILIC; injection-solvent strength and peak shape; repeatable partial equilibration in gradient mode; ≈3× ESI-MS sensitivity gain under stated conditions.
Taylor T. The LCGC Blog: Hydrophilic interaction chromatography — practical insights. LCGC International. 2017. https://www.chromatographyonline.com/view/lcgc-blog-hydrophilic-interaction-chromatography-practical-insights — practitioner heuristics for HILIC equilibration volumes and sample diluent water content; diluent mismatch symptoms; needle-wash solvent.
Stoll DR, Seidl C. Column re-equilibration following gradient elution: how long is long enough? Part I: reversed-phase and HILIC separations of small molecules. LCGC North America. 2019;37(11). https://www.chromatographyonline.com/view/column-re-equilibration-following-gradient-elution-how-long-long-enough-part-i-reversed-phase-and-hi — partial vs full re-equilibration in gradient HILIC; repeatable results with short, precisely held re-equilibration; recommendation to vary re-equilibration time during development.
United States Pharmacopeial Convention. General Chapter ⟨621⟩ Chromatography, Stage 4 Harmonization (official 1 December 2022). https://www.usp.org/sites/default/files/usp/document/harmonization/gen-chapter/harmonization-november-2021-m99380.pdf (accessed 2026-10-16) — allowable adjustments exclude changing the stationary-phase substituent; changes beyond them require revalidation; gradient criterion that first peaks be sufficiently retained (no numeric k); system suitability throughout.

When Should You Switch from Reversed-Phase HPLC to HILIC?