pKa Calculations

CHEMSMART provides pKa workflows in two separate stages:

Job submission — generate and run Gaussian or ORCA calculations for HA, A⁻, and (optionally) a reference acid. See Gaussian pKa Calculations and ORCA pKa Calculations.
Output analysis — compute pKa values from completed output files using the backend-independent command chemsmart run pka. Analysis is program-agnostic: the same workflow reads Gaussian .log and ORCA .out files and extracts the energies and thermal corrections needed for the pKa cycle.

pKa Calculations

Execution Architecture 

Job submission

chemsmart run/sub gaussian ... pka [submit|batch] — prepare and run Gaussian pKa calculations.
chemsmart run/sub orca ... pka [submit|batch] — prepare and run ORCA pKa calculations.
Use chemsmart run for local preparation and execution; use chemsmart sub on HPC clusters to generate scheduler scripts (see HPC Cluster Submission (chemsmart sub)).
A single structure yields one job; batch input (CSV table or multi-molecule CDXML) can produce multiple jobs in one invocation.
When pka is invoked without an explicit subcommand, a submission table triggers batch; otherwise submit runs.

Output analysis

chemsmart run pka analyze — single-system analysis from up to eight output files.
chemsmart run pka batch-analyze — table-driven batch analysis.
Both commands use the same pKa analysis workflow. Gaussian .log and ORCA .out files can be mixed in the same batch table; each file is read and interpreted on its own.
Analysis never invokes gaussian or orca job submission — only reads completed output files.

Theory 

The pKa of an acid HA in aqueous solution is defined by the equilibrium:

\[\text{HA}_{(\text{aq})} \rightleftharpoons \text{A}^{-}_{(\text{aq})} + \text{H}^{+}_{(\text{aq})}\]

The pKa is related to the standard Gibbs free energy change by:

\[\text{p}K_{\text{a}} = \frac{\Delta G^{\circ}_{\text{aq}}}{2.303 \cdot R \cdot T}\]

where \(R\) is the gas constant and \(T\) is the temperature.

Thermodynamic Cycles 

CHEMSMART supports two thermodynamic cycles for pKa calculations:

1. Proton Exchange (Isodesmic) Cycle (Default, Recommended)

This is the default when -s is omitted for both job submission and output analysis. A reference acid HRef with known experimental pKa is required to cancel systematic errors:

\[\text{HA} + \text{Ref}^{-} \rightarrow \text{A}^{-} + \text{HRef}\]

The pKa is computed as:

\[\text{p}K_{\text{a}}(\text{HA}) = \text{p}K_{\text{a}}(\text{HRef}) + \frac{\Delta G_{\text{soln}}}{2.303 \cdot R \cdot T}\]

where:

\[\Delta G_{\text{soln}} = \left[ G(\text{A}^{-})_{\text{soln}} + G(\text{HRef})_{\text{soln}} \right] - \left[ G(\text{HA})_{\text{soln}} + G(\text{Ref}^{-})_{\text{soln}} \right]\]

2. Direct Cycle

Uses the absolute free energy of a proton in water:

\[\text{p}K_{\text{a}} = \frac{G(\text{A}^{-})_{\text{aq}} - G(\text{HA})_{\text{aq}} + \Delta G^{\circ}(\text{H}^{+})_{\text{aq}}}{2.303 \cdot R \cdot T}\]

Default value: \(\Delta G^{\circ}(\text{H}^{+})_{\text{aq}} = -265.9\) kcal/mol (Tissandier et al., 1998).

Dual-Level Approach 

CHEMSMART implements a dual-level approach for accurate solvation free energies:

Thermal corrections (\(G_{\text{corr}}\)) from gas-phase frequency calculations using quasi-harmonic Gibbs free energy:

\[G_{\text{corr}} = G_{\text{qh}}(T) - E_{\text{gas}}\]
Solvent energies (\(E_{\text{solv}}\)) from high-level single-point calculations in implicit solvent (e.g., SMD or CPCM).
Total free energy in solution:

\[G_{\text{soln}} = E_{\text{solv}} + G_{\text{corr}}\]

Note

All internal energies are stored in Hartree (au). \(\Delta G_{\text{soln}}\) (proton exchange) and \(\Delta G_{\text{diss}}\) (direct dissociation) are converted to kcal/mol for the pKa formula (1 Hartree = 627.5094740631 kcal/mol).

Job Submission (Gaussian / ORCA)

Job submission is backend-specific. Use the dedicated pages for full examples and parameter tables:

Commands

The default scheme is proton exchange, which requires a reference acid (-r, -rpi, -rc, -rm). Use -s direct only when you want the direct dissociation cycle without a reference acid.

# Proton exchange (default) — reference acid required
chemsmart run gaussian -p my_project -f acid.xyz -c 0 -m 1 pka \
    -pi 10 -r ref_acid.xyz -rpi 21 -rc 1 -rm 1

chemsmart run orca -p my_project -f acid.xyz -c 0 -m 1 pka \
    -pi 10 -r ref_acid.xyz -rpi 21 -rc 1 -rm 1

# Direct cycle — no reference acid; must set -s direct explicitly
chemsmart run gaussian -p my_project -f acid.xyz -c 0 -m 1 pka -pi 10 -s direct
chemsmart run orca -p my_project -f acid.xyz -c 0 -m 1 pka -pi 10 -s direct

# Batch submission (proton exchange requires reference options on the pka group)
chemsmart run gaussian -p my_project -f pka_input.csv pka \
    -r ref_acid.xyz -rpi 21 -rc 1 -rm 1 batch

# Batch submission (direct cycle)
chemsmart run gaussian -p my_project -f pka_input.csv pka -s direct batch

Submission input table (pka batch)

Comma- or whitespace-delimited table with columns filepath, proton_index, charge, multiplicity.

ChemDraw CDXML / CDX Input (pKa Jobs)

pKa job submission can read structures directly from ChemDraw .cdxml and .cdx files. CHEMSMART reads atom colours in the drawing to identify the acidic proton to remove. Colour the proton (or the H in a functional group such as –OH) in ChemDraw with a distinct colour; CHEMSMART auto-detects it so -pi is often unnecessary.

Single-molecule submit

When -f points to one CDXML structure with a single fragment, omit -pi if the coloured proton is unique:

chemsmart run gaussian -p my_project -f phenol.cdxml -c 0 -m 1 pka \
    -r ref_acid.xyz -rpi 21 -rc 1 -rm 1

chemsmart run gaussian -p my_project -f phenol.cdxml -c 0 -m 1 pka -s direct

Use -cc / --color-code when several hydrogens share similar styling and you need to select a specific ChemDraw colour-table index. The reference acid may also be a CDXML file; in that case -rpi can be omitted when the reference proton is uniquely coloured (or use -rcc / --reference-color-code).

Multi-molecule CDXML (one job per fragment)

A single .cdxml / .cdx file may contain multiple molecules (multiple ChemDraw fragments). CHEMSMART performs per-fragment coloured-proton detection and creates one pKa job per fragment.

Pass the file with pka batch (or pka submit for a single-fragment file):

chemsmart run gaussian -p my_project -f acids.cdxml -c 0 -m 1 pka \
    -r ref_acid.xyz -rpi 21 -rc 1 -rm 1 batch

chemsmart run orca -p my_project -f acids.cdxml -c 0 -m 1 pka -s direct batch

Job labels are derived from the filename, e.g. acids_frag1_pka (Gaussian) or acids_frag1_pka (ORCA).

Charge and multiplicity

How charge and multiplicity are resolved depends on the input mode:

Input mode	Charge / multiplicity source
CSV batch table	Required columns on each row (`charge`, `multiplicity`). The parent `gaussian` / `orca` command does not need `-c` / `-m` when `-f` is a table.
Multi-fragment CDXML (`-f` is `.cdxml` / `.cdx`)	Parent `-c` / `-m` apply to every fragment by default. If either is omitted on the backend command, CHEMSMART may copy values from the parsed CDXML structure when the drawing supplies them.
Single-molecule submit (XYZ, LOG, CDXML, …)	`-c` and `-m` on the backend command are required unless already present on merged project/job settings.

Blank proton_index in a single-molecule CDXML table row triggers coloured-proton auto-detection only; it does not remove the requirement for charge and multiplicity columns in CSV tables.

CDXML paths inside a CSV batch table

You can mix XYZ and CDXML inputs in the same submission table. Each row becomes one pKa job. For a single-molecule .cdxml / .cdx row, leave proton_index blank to auto-detect the coloured proton; an explicit value overrides the detection (same behaviour as single-file CDXML submit). Non-CDXML rows still require proton_index.

filepath,proton_index,charge,multiplicity
/path/to/acid1.xyz,12,0,1
/path/to/acid2.cdxml,,0,1

Multi-molecule CDXML files cannot be expanded from a table row. Pass them directly as -f with pka batch (see above) to create one job per ChemDraw fragment.

Note

If -f is a CDXML file (not a CSV table), CHEMSMART routes to coloured-proton batch expansion automatically. For general CDXML structure handling outside pKa, see ChemDraw Files.

Proton and reference options for CDXML

Short	Long	Description
`-pi`	`--proton-index`	Optional for CDXML when a uniquely coloured proton is present. Required for XYZ/LOG/COM inputs.
`-cc`	`--color-code`	ChemDraw colour-table index for the target acidic proton (`.cdxml` / `.cdx` only).
`-rpi`	`--reference-proton-index`	Optional when `-r` is a CDXML file with a uniquely coloured reference proton.
`-rcc`	`--reference-color-code`	ChemDraw colour-table index for the reference proton (`.cdxml` / `.cdx` reference only).

Job output file naming

Each pKa job creates gas-phase opt+freq and solvent single-point sub-jobs. Output filenames follow the sub-job label:

Sub-job label suffix	Species
`_HA_opt`	Target acid HA (gas-phase opt+freq)
`_A_opt`	Target conjugate base A⁻ (gas-phase opt+freq)
`_HA_sp`	HA solvent single-point
`_A_sp`	A⁻ solvent single-point
`_HRef_opt` / `_Ref_opt`	Reference acid / conjugate base (proton exchange only)
`_HRef_sp` / `_Ref_sp`	Reference solvent single-points (proton exchange only)

Gaussian batch jobs use the input stem as the job label (e.g. acid1_HA_opt.log). ORCA batch jobs append _pka to the stem (e.g. acid1_pka_HA_opt.out). The output-analysis autodiscovery convention below is aligned with the {basename}_pka_* pattern used by ORCA submission and by typical batch output tables.

HPC Cluster Submission (`chemsmart sub`)

On a cluster, use chemsmart sub instead of chemsmart run to write scheduler scripts and per-job run wrappers. The pKa workflow is unchanged at the chemistry level; only the launch path differs.

Single job

chemsmart sub gaussian -p my_project -f acid.xyz -c 0 -m 1 pka -pi 10 -s direct submit

Batch table or multi-fragment CDXML

One scheduler submission is created per table row or per ChemDraw fragment. CHEMSMART expands pka batch into multiple jobs locally, then writes a separate chemsmart_sub_<label>.sh and chemsmart_run_<label>.py for each job.

chemsmart sub gaussian -p my_project -f pka_input.csv pka -s direct batch

chemsmart sub gaussian -p my_project -f acids.cdxml -c 0 -m 1 pka -s direct batch

Per-job script reconstruction 

Each cluster run wrapper must replay one pKa submission, not the entire batch table or full multi-fragment CDXML file. When a job is created from pka batch, CHEMSMART stores row- or fragment-level metadata and rewrites the CLI inside chemsmart_run_<label>.py before submission:

CSV batch rows — replace the table path in -f / --filename with that row’s filepath; change batch to submit.
Multi-fragment CDXML — point -f at the same CDXML file but add --index / -i so only one fragment is processed; change batch to submit.
Explicit per-job options — inject or update --proton-index, --charge, --multiplicity, and --label so the reconstructed command is self-contained and passes Click validation on the cluster node.
Proton exchange tables — rows after the first may run with -s direct; reference-acid flags are dropped from later rows automatically (same behaviour as local pka batch).

Example: a two-row CSV batch submitted with chemsmart sub ... pka batch yields two run scripts. The script for row two might equivalent to:

chemsmart run gaussian -p my_project -f /path/to/acid2.xyz -c 0 -m 1 \
    pka -pi 8 -s direct submit

Example: a five-fragment CDXML file pka_scale.cdxml yields labels such as pka_scale_frag1_pka, …, pka_scale_frag5_pka. Each run script targets one fragment via --index and the matching --proton-index, --label, -c, and -m.

This reconstruction is what allows chemsmart sub ... pka batch on a cluster to behave like five independent pka submit calls while you only maintain one top-level submission command locally.

See also Command Line Interface Overview for general chemsmart sub usage and Server Settings for scheduler configuration.

Output Analysis (`chemsmart run pka`)

All post-processing lives under chemsmart run pka. No Gaussian or ORCA backend is invoked during analysis.

Thermochemistry extraction 

For each output file, analysis:

Open the file and detect the program (Gaussian or ORCA).
Reads the gas-phase SCF energy and quasi-harmonic Gibbs free energy (for opt+freq outputs).
Reads the solvent-phase SCF energy (for single-point outputs).
Raises a clear error if a required quantity cannot be extracted.

Computing pKa from Output Files (`analyze`)

If you have completed output files for a single acid, compute pKa with analyze.

Proton exchange (default)

Only -ha and -hr are strictly required; the remaining six companion files are auto-discovered when they follow the naming convention below. -rp / --reference-pka is required.

chemsmart run pka analyze \
    -ha acid1_pka_HA_opt.log \
    -hr ref_acid_pka_HRef_opt.log \
    -rp 6.75 \
    -T 333.15 -c 1.0 -csg 100 -ch 100

Provide all eight files explicitly when auto-discovery is not appropriate:

chemsmart run pka analyze \
    -ha acid1_pka_HA_opt.log \
    -a acid1_pka_A_opt.log \
    -hr ref_acid_pka_HRef_opt.log \
    -r ref_acid_pka_Ref_opt.log \
    -has acid1_pka_HA_sp.log \
    -as acid1_pka_A_sp.log \
    -hrs ref_acid_pka_HRef_sp.log \
    -rs ref_acid_pka_Ref_sp.log \
    -rp 6.75 \
    -T 298.15

Direct dissociation

Four output files are required (HA, A⁻, and their solvent single-points). Both -s direct and -dG must be specified on the pka group before the analyze subcommand:

chemsmart run pka -s direct -dG -265.9 analyze \
    -ha acid1_pka_HA_opt.log \
    -T 298.15

Only -ha is strictly required; -a, -has, and -as are auto-discovered from the target-acid suffix convention when omitted.

File autodetection (`analyze`)

When companion paths are omitted, CHEMSMART derives them from the HA and HRef gas-phase files using the same suffix patterns as batch-analyze:

From the HA gas-phase file (``-ha``)

<basename>_pka_A_opt.<ext> — conjugate base gas-phase
<basename>_pka_HA_sp.<ext> — HA solvent single-point
<basename>_pka_A_sp.<ext> — conjugate base solvent SP

From the HRef gas-phase file (``-hr``)

<basename>_pka_Ref_opt.<ext> — reference conjugate base
<basename>_pka_HRef_sp.<ext> — reference acid solvent SP
<basename>_pka_Ref_sp.<ext> — reference conjugate base solvent SP

Alternative suffixes (_HRef_opt, _pka_cb, etc.) are also recognised. The file extension (.log or .out) is chosen from the detected program. Override any auto-discovered path with the corresponding flag.

If a required file is missing or cannot be parsed, analysis stops with a clear error (missing paths or missing thermochemistry data).

Batch Processing of Output Files (`batch-analyze`)

Parse a table of pre-computed output file paths to calculate pKa values in batch.

Proton exchange (default)

chemsmart run pka -T 333.15 -c 1.0 -csg 100 -ch 100 batch-analyze \
    -o pka_output_table.csv \
    -O results.dat

Direct dissociation

Both -s direct and -dG are required:

chemsmart run pka -s direct -dG -265.9 batch-analyze \
    -o pka_output_table_direct.csv \
    -O results_direct.dat

The formatted batch summary table is printed to stdout. When -O / --output-results is given, the same formatted report is written to that file (not a wide CSV of input columns).

Output table format

The output table (-o / --output-table) must contain at least a basename column. Other file paths may be omitted and are auto-discovered from basename when blank.

Required column

basename: Unique identifier for each acid. Used for file auto-discovery.

Target-acid columns (both schemes)

Accepted header aliases include ha_opt, a_opt, ha_solv, a_solv, etc. (see column list below).

When blank, CHEMSMART searches for <basename><suffix>.<ext> in the current working directory. Suffixes are tried in order; both .log and .out are tested.

Column	Description	Auto-discovery suffixes (first match wins)
`ha_gas`	HA gas-phase opt+freq output	`_pka_HA_opt`, `_pka_HA`, `_pka`
`a_gas`	A⁻ gas-phase opt+freq output	`_pka_A_opt`, `_pka_A`, `_pka_cb`
`ha_sp`	HA solvent single-point output	`_pka_HA_sp`, `_pka_sp`
`a_sp`	A⁻ solvent single-point output	`_pka_A_sp`, `_pka_cb_sp`

Reference-acid columns (proton exchange only)

Ignored for direct dissociation. Blank reference columns inherit values from the previous row.

Column	Description
`href_gas`	HRef gas-phase opt+freq output (not auto-discovered from `basename`; provide explicitly or inherit)
`ref_gas`	Ref⁻ gas-phase opt+freq output
`href_sp`	HRef solvent single-point output
`ref_sp`	Ref⁻ solvent single-point output
`pka_ref`	Experimental pKa of the reference acid

Example pka_output_table.csv:

basename,ha_gas,a_gas,ha_sp,a_sp,href_gas,ref_gas,href_sp,ref_sp,pka_ref
phenol,,,,,ref_acid_pka_HRef_opt.log,ref_acid_pka_Ref_opt.log,ref_acid_pka_HRef_sp.log,ref_acid_pka_Ref_sp.log,6.75
benzoic_acid,,,,,,,,,6.75

`batch-analyze` options

Short	Long	Description
`-o`	`--output-table`	Required. Path to the output-file table.
`-O`	`--output-results`	Optional path for the formatted results report. Stdout always receives the summary table.
`-p`	`--program`	Require every populated output path to match `gaussian` or `orca`. Default: `auto` (per-file detection; supports mixed tables).

Mixed Gaussian / ORCA tables

With -p auto (default), the program behind each output file is detected automatically. A batch table may contain Gaussian .log targets and ORCA .out reference files in the same run. Use -p gaussian or -p orca only when you want to validate that all populated paths belong to one backend.

Analysis Scheme Options 

These options apply to chemsmart run pka (analyze and batch-analyze). They are separate from submission options on chemsmart run/sub gaussian ... pka and chemsmart run/sub orca ... pka.

Short	Long	Description
`-s`	`--scheme`	Thermodynamic cycle: `direct` or `proton exchange`. Default: `proton exchange`.
`-dG`	`--delta-g-proton`	\(G_{\text{soln}}(\text{H}^{+})\) in kcal/mol for the direct cycle. Required when `-s direct` is used for analysis. Has no default during analysis.

Note

For job submission, -dG defaults to -265.9 kcal/mol. For output analysis you must pass -dG explicitly whenever -s direct is used.

Output File Options 

Used by analyze (not batch-analyze, which reads paths from the table).

Gas-phase optimization + frequency files

Short	Long	Description
`-ha`	`--ha`	HA gas-phase opt+freq output.
`-a`	`--a`	A⁻ gas-phase opt+freq output.
`-hr`	`--href`	HRef gas-phase opt+freq output.
`-r`	`--ref`	Ref⁻ gas-phase opt+freq output.

Solvent single-point files

Short	Long	Description
`-has`	`--ha-solv`	HA solvent single-point output.
`-as`	`--a-solv`	A⁻ solvent single-point output.
`-hrs`	`--href-solv`	HRef solvent single-point output.
`-rs`	`--ref-solv`	Ref⁻ solvent single-point output.

Thermochemistry Options 

Shared by analyze and batch-analyze.

Short	Long	Description
`-T`	`--temperature`	Temperature in Kelvin. Default: `298.15` K.
`-c`	`--concentration`	Concentration in mol/L. Default: `1.0` mol/L.
`-P`	`--pressure`	Pressure in atm. Default: `1.0` atm.
`-csg`	`--cutoff-entropy-grimme`	Cutoff frequency (cm⁻¹) for entropy using Grimme’s quasi-RRHO. Default: `100.0`.
`-cst`	`--cutoff-entropy-truhlar`	Cutoff frequency (cm⁻¹) for entropy using Truhlar’s quasi-RRHO. Mutually exclusive with `-csg`.
`-ch`	`--cutoff-enthalpy`	Cutoff frequency (cm⁻¹) for enthalpy using Head-Gordon’s method. Default: `100.0`.
`-rp`	`--reference-pka`	Experimental pKa of HRef. Required for proton exchange analysis.

Output Format 

When computing pKa from output files, CHEMSMART prints a detailed summary. The format depends on the analysis scheme.

Proton exchange

==============================================================================
pKa Calculation - Dual-level Proton Exchange Scheme
==============================================================================
Reaction: HA + Ref- -> A- + HRef
Temperature: 373.15 K

Method:
  G_corr = qh-G(T) - E_gas  (from gas-phase freq calculation)
  G_soln = E_solv + G_corr  (solution free energy)
  DG_soln = [G(A-)_soln + G(HRef)_soln] - [G(HA)_soln + G(Ref-)_soln]
  pKa = pKa_ref + DG_soln / (RT * ln10)
------------------------------------------------------------------------------

Gas-Phase Electronic Energies (E_gas, au):
  HA:    -345.7419436500
  A-:    -344.9153986020
  HRef:  -365.8436493070
  Ref-:  -365.4561783660

Thermal Corrections (G_corr = qh-G - E_gas, au):
  HA:    0.0931931305
  A-:    0.0758935969
  HRef:  0.1404528844
  Ref-:  0.1267467582

Solvent Single-Point Energies (E_solv, au):
  HA:    -346.4882221850
  A-:    -345.8989956310
  HRef:  -366.5974351550
  Ref-:  -366.1368369100

Solution Free Energies (G_soln = E_solv + G_corr, au):
  HA:    -346.3950290545
  A-:    -345.8231020341
  HRef:  -366.4569822706
  Ref-:  -366.0100901518
------------------------------------------------------------------------------

pKa Calculation:
  DG_soln = 0.1250349015 au
          = 78.4606 kcal/mol
  pKa(HRef)_ref = 6.75

  *** Computed pKa(HA) = 52.70 ***
==============================================================================

Direct dissociation

==============================================================================
pKa Calculation - Direct Dissociation Scheme
==============================================================================
Reaction: HA -> A- + H+
Temperature: 298.15 K

Method:
  G_corr = qh-G(T) - E_gas  (from gas-phase freq calculation)
  G_soln = E_solv + G_corr  (solution free energy)
  DG_diss = G_soln(A-) + G_soln(H+) - G_soln(HA)
  pKa = DG_diss / (2.303 * R * T)
------------------------------------------------------------------------------

Gas-Phase Electronic Energies (E_gas, au):
  HA:  -345.7419436500
  A-:  -344.9153986020

Thermal Corrections (G_corr = qh-G - E_gas, au):
  HA:  0.0931931305
  A-:  0.0758935969

Solvent Single-Point Energies (E_solv, au):
  HA:  -346.4882221850
  A-:  -345.8989956310

Solution Free Energies (G_soln = E_solv + G_corr, au):
  HA:  -346.3950290545
  A-:  -345.8231020341
------------------------------------------------------------------------------

pKa Calculation:
  G_soln(H+) = -265.9000 kcal/mol
  DG_diss = 0.1250349015 au
          = 78.4606 kcal/mol

  *** Computed pKa(HA) = 52.70 ***
==============================================================================

Batch analyze output

batch-analyze prints a compact table whose ΔG column header matches the scheme. The same formatted report is written to -O when provided.

Typical table (single-acid basenames):

==============================================================================
Batch pKa Results (Dual-level Proton Exchange)
==============================================================================
Temperature: 298.15 K
Pressure: 1.0 atm
basename                              pKa   ΔG_soln (kcal/mol)
------------------------------------------------------------------------------
phenol                               10.12             13.4567
benzoic_acid                          4.20              5.7890
==============================================================================

Multi-fragment CDXML workflow — after chemsmart sub ... -f pka_scale.cdxml pka batch and chemsmart run pka batch-analyze, basenames match the fragment labels (<stem>_frag<N>_pka or the Gaussian stem without _pka suffix, depending on backend). Example (values from a test reference acid; not physically meaningful):

==============================================================================
Batch pKa Results (Dual-level Proton Exchange)
==============================================================================
Temperature: 278.15 K
Pressure: 1.0 atm
basename                              pKa   ΔG_soln (kcal/mol)
------------------------------------------------------------------------------
pka_scale_frag1                      -8.17            -23.8913
pka_scale_frag4                      -1.24            -15.0654
pka_scale_frag3                       9.22             -1.7521
pka_scale_frag5                      -2.68            -16.8958
==============================================================================

For direct dissociation, the header reads Batch pKa Results (Direct Dissociation) and the column is labeled ΔG_diss (kcal/mol).

References 

Tissandier, M. D. et al. (1998). J. Phys. Chem. A, 102, 7787. (Absolute proton solvation energy)
Grimme, S. (2012). Chem. Eur. J., 18, 9955. (Quasi-RRHO method)
Marenich, A. V.; Cramer, C. J.; Truhlar, D. G. (2009). J. Phys. Chem. B, 113, 6378. (SMD solvation model)

pKa Calculations

Output table format

batch-analyze options

Mixed Gaussian / ORCA tables

`batch-analyze` options