Titration - Acid-Base Analysis

Acid-base potentiometric titration is used to measure freebase/limestone content in pilot plant liquids. The OrionStarT940 Auto Titrator generates pH vs. volume curves and calculates endpoint values.

Method Overview

Property	Value
Full Name	Acid-Base Potentiometric Titration
Purpose	Measures freebase/limestone content in pilot plant liquids
Instrument	OrionStarT940 Auto Titrator
Output	pH vs. volume titration curves and endpoint values
SharePoint Location	Analytical Data > OrionStarT940 Auto Titrator > Data Exports
File Format	CSV (paired files)

Instrument

OrionStarT940 Auto Titrator

The OrionStarT940 is an automated potentiometric titrator that measures pH as a function of titrant volume added. It automatically detects endpoints based on inflection points in the pH curve.

Output Files:

• *_Tsum.csv - Summary/metadata file (sample ID, method, endpoint values)
• *_Tdata1.csv - Data file (pH vs. volume curve points)

Sample Types

Type	Description
BBE Pilot Liquids	Liquid samples from the BBE pilot plant

Sample ID Format

Titrator sample IDs follow pilot plant naming conventions:

Input Format	Canonical Sample ID
R204121620	PIL13-441-121620-R204-LIQ-RUSH
D101121620	PIL13-441-121620-D101-LIQ-RUSH
121700D101	PIL13-441-121700-D101-LIQ-RUSH

The format is: PIL13-441-{MMDDYY}-{Location}-LIQ-RUSH

Data Pipeline

SharePoint Source

• Site: Analytical Data
• Folder: OrionStarT940 Auto Titrator/Data Exports/
• File Type: CSV (paired files)

File Pairing

Files are generated in pairs with matching prefixes:

T40277_20251216_1711_506_Tsum.csv   <- Metadata
T40277_20251216_1711_506_Tdata1.csv <- Curve data

The pipeline automatically:

1. Separates files by type (Tsum vs Tdata1)
2. Matches pairs by prefix
3. Fetches missing pairs from SharePoint when only one file arrives

Dagster Assets

The titration data pipeline is defined in apps/datasmart/src/datasmart/assets/analytical/tritation.py:

freebase_titration (sharepoint_multi_asset)
├── freebase_titration_samples (analytical.freebase_titration_samples)
└── freebase_titration_data (analytical.freebase_titration_data)

Asset Description

Asset	Schema	Description
freebase_titration_samples	analytical	Sample metadata and endpoint values
freebase_titration_data	analytical	pH vs volume curve data points

Database Tables

analytical.freebase_titration_samples

Metadata table with one row per sample.

Column	Type	Description
file_id	string	SharePoint file identifier (primary key)
original_sample_id	string	Raw sample ID from file
sample_id	string	Normalized pilot sample ID
sample_date	datetime	Analysis date/time
titrator_name	string	Titrator instrument name
method_name	string	Titration method
titrant_name	string	Titrant solution name
titrant_conc	float	Titrant concentration (M)
sample_amount	float	Sample amount (g or mL)
sample_conc	float	Sample concentration
ep_volume	float	Endpoint volume (mL)
ep_value	float	Endpoint pH value
initial_ph	float	Initial pH
temperature	float	Temperature (C)
duration	string	Analysis duration (min)
sp_site	string	SharePoint site
file_path	string	File path
file_url	string	SharePoint URL
last_update	datetime	Last modification time

analytical.freebase_titration_data

Curve data table with multiple rows per sample (one per data point).

Column	Type	Description
file_id	string	SharePoint file identifier
volume	float	Titrant volume (mL)
ph	float	pH value

File Formats

Tsum File (Metadata)

Comma-separated key-value pairs:

Titrator Name, OrionStarT940
Date, 12/16/2024
Time, 05:11 PM
Method Name, Freebase
Sample ID, R204121620
Titrant Name, 0.1M HCl
Titrant Conc., 0.1 M
Sample Amount, 10.000 gram
EP Volume 1, 15.234 mL
EP Value 1, 4.50
Initial pH, 12.34
Temperature, 25.0
Duration (min), 5:30

Tdata1 File (Curve Data)

CSV with header row followed by data:

Point,Volume (mL),E(pH),dE/dV,d2E/dV2
1,0.000,12.34,0.00,0.00
2,0.100,12.30,-0.40,0.10
3,0.200,12.25,-0.50,0.10
...

Usage Examples

Query Sample Metadata

from shared.db.sql import SQL

# Get all samples with endpoints
samples = SQL.read("""
    SELECT sample_id, sample_date, ep_volume, ep_value, initial_ph
    FROM analytical.freebase_titration_samples
    WHERE sample_id IS NOT NULL
    ORDER BY sample_date DESC
""")

Get Titration Curve

# Get curve for a specific sample
curve = SQL.read("""
    SELECT d.volume, d.ph
    FROM analytical.freebase_titration_data d
    JOIN analytical.freebase_titration_samples s ON d.file_id = s.file_id
    WHERE s.sample_id = 'PIL13-441-121620-R204-LIQ-RUSH'
    ORDER BY d.volume
""")

Filter by Location

# Get all samples from R204 location
r204_samples = SQL.read("""
    SELECT *
    FROM analytical.freebase_titration_samples
    WHERE sample_id LIKE 'PIL13-441-%-R204-%'
""")

Plot Titration Curve

import matplotlib.pyplot as plt
from shared.db.sql import SQL

# Get curve data
curve = SQL.read("""
    SELECT d.volume, d.ph
    FROM analytical.freebase_titration_data d
    JOIN analytical.freebase_titration_samples s ON d.file_id = s.file_id
    WHERE s.sample_id = 'PIL13-441-121620-R204-LIQ-RUSH'
    ORDER BY d.volume
""")

# Get endpoint values
sample = SQL.read("""
    SELECT ep_volume, ep_value
    FROM analytical.freebase_titration_samples
    WHERE sample_id = 'PIL13-441-121620-R204-LIQ-RUSH'
""").iloc[0]

plt.figure(figsize=(10, 6))
plt.plot(curve['volume'], curve['ph'], 'b-', linewidth=1.5)
plt.axhline(y=sample['ep_value'], color='r', linestyle='--', label=f"Endpoint: pH {sample['ep_value']}")
plt.axvline(x=sample['ep_volume'], color='g', linestyle='--', label=f"EP Volume: {sample['ep_volume']} mL")
plt.xlabel('Volume (mL)')
plt.ylabel('pH')
plt.title('Titration Curve: PIL13-441-121620-R204-LIQ-RUSH')
plt.legend()
plt.grid(True, alpha=0.3)
plt.show()

Sample ID Parsing

The pipeline automatically converts raw titrator sample IDs to canonical pilot plant format:

import re

SAMPLE_ID_LOCATION_DATE_REGEX = re.compile(r"^([A-Z]\d{3})(\d{6})$")
SAMPLE_ID_DATE_LOCATION_REGEX = re.compile(r"^(\d{6})([A-Z]\d{3})$")

def parse_pilot_sample_id(raw_sample_id: str) -> str | None:
    """
    Parse a titrator sample ID into canonical pilot sample ID format.

    Examples:
        R204121620 -> PIL13-441-121620-R204-LIQ-RUSH
        D101121620 -> PIL13-441-121620-D101-LIQ-RUSH
        121700D101 -> PIL13-441-121700-D101-LIQ-RUSH
    """
    raw_sample_id = raw_sample_id.strip().upper()

    # Try location + date format (e.g., R204121620)
    match = SAMPLE_ID_LOCATION_DATE_REGEX.match(raw_sample_id)
    if match:
        location = match.group(1)
        date = match.group(2)
        return f"PIL13-441-{date}-{location}-LIQ-RUSH"

    # Try date + location format (e.g., 121700D101)
    match = SAMPLE_ID_DATE_LOCATION_REGEX.match(raw_sample_id)
    if match:
        date = match.group(1)
        location = match.group(2)
        return f"PIL13-441-{date}-{location}-LIQ-RUSH"

    return None

Related Constants

TITRATOR_FOLDER = "OrionStarT940 Auto Titrator/Data Exports"
TITRATOR_SITE = "Analytical Data"

Troubleshooting

Missing Sample ID

If sample_id is NULL:

1. The raw sample ID didn't match expected pilot plant formats
2. Check original_sample_id column for the raw value
3. Verify the sample ID follows LNNN + MMDDYY or MMDDYY + LNNN format

Missing Curve Data

If curve data is empty:

1. Check if the paired Tdata1 file exists in SharePoint
2. Verify the file prefix matches between Tsum and Tdata1 files
3. Check Dagster logs for parsing errors

Unpaired Files

The pipeline attempts to fetch missing pairs from SharePoint. If files remain unpaired:

1. Verify both files exist in the SharePoint folder
2. Check that filenames have matching prefixes (before _Tsum.csv / _Tdata1.csv)

Source Files

• apps/datasmart/src/datasmart/assets/analytical/tritation.py - Dagster asset definitions