Hazard Evaluation¶

PHOENIX implements CHETAH-style hazard screening based on ASTM E659.

Screening Only

Hazard evaluation is for screening purposes only. Always validate results experimentally before handling potentially hazardous materials.

Quick Evaluation¶

from phoenix import Compound

compound = Compound.from_smiles("c1ccccc1[N+](=O)[O-]")  # Nitrobenzene
result = compound.evaluate_hazard()

print(f"Hazard Class: {result.hazard_class}")
print(f"Triggered Criteria: {result.triggered_criteria}")

HazardResult¶

The evaluate_hazard() method returns a HazardResult dataclass:

Attributes¶

Attribute	Type	Description
`smiles`	str	Canonical SMILES
`formula`	str	Molecular formula
`molecular_weight`	float	MW in g/mol
`delta_hf_kJ_mol`	float	ΔHf° in kJ/mol
`max_decomposition_kJ_mol`	float	ΔHd in kJ/mol
`max_decomposition_cal_g`	float	ΔHd in cal/g
`oxygen_balance_percent`	float	OB%
`hazard_class`	str	HIGH, MEDIUM, or LOW
`triggered_criteria`	tuple[int]	List of triggered criteria (1-4)
`functional_group_alerts`	tuple[str]	Names of detected groups
`product_breakdown`	dict	Decomposition products
`gas_volume_L_g`	float	Gas generation in L/g

Example¶

result = compound.evaluate_hazard()

# Access all properties
print(f"Formula: {result.formula}")
print(f"MW: {result.molecular_weight:.2f} g/mol")
print(f"ΔHf°: {result.delta_hf_kJ_mol:.1f} kJ/mol")
print(f"ΔHd: {result.max_decomposition_cal_g:.1f} cal/g")
print(f"OB%: {result.oxygen_balance_percent:.1f}%")
print(f"Hazard Class: {result.hazard_class}")
print(f"Criteria: {result.triggered_criteria}")
print(f"Alerts: {result.functional_group_alerts}")
print(f"Gas: {result.gas_volume_L_g:.3f} L/g")

CHETAH Criteria¶

Criterion 1: High Instability¶

ΔHd < -300 cal/g

Compounds with very high decomposition energy.

# Example: Check for criterion 1
if result.max_decomposition_cal_g < -300:
    print("HIGH hazard: Criterion 1 triggered")

Criterion 2: Medium Instability¶

-300 ≤ ΔHd < -100 cal/g

Compounds with moderate decomposition energy.

# Example: Check for criterion 2
if -300 <= result.max_decomposition_cal_g < -100:
    print("MEDIUM hazard: Criterion 2 triggered")

Criterion 3: Oxygen Balance¶

-200% < OB% < +100%

Compounds with reactive oxygen balance.

# Example: Check for criterion 3
if -200 < result.oxygen_balance_percent < 100:
    print("OB% in reactive range: Criterion 3 triggered")

Criterion 4: Functional Groups¶

Presence of known reactive functional groups.

# Example: Check for criterion 4
if result.functional_group_alerts:
    print("Functional groups detected: Criterion 4 triggered")
    for alert in result.functional_group_alerts:
        print(f"  - {alert}")

Hazard Classification Logic¶

IF Criterion 1 triggered:
    → HIGH

ELSE IF Criterion 2 triggered:
    → MEDIUM

ELSE IF Criteria 3 AND 4 triggered:
    → MEDIUM

ELSE IF ANY criterion triggered:
    → MEDIUM

ELSE:
    → LOW

Examples by Hazard Class¶

HIGH Hazard¶

# TNT - Trinitrotoluene
tnt = Compound.from_smiles("Cc1c([N+](=O)[O-])cc([N+](=O)[O-])cc1[N+](=O)[O-]")
result = tnt.evaluate_hazard()
print(f"TNT: {result.hazard_class}")  # HIGH
print(f"ΔHd = {result.max_decomposition_cal_g:.0f} cal/g")

MEDIUM Hazard¶

# Nitrobenzene
nitrobenzene = Compound.from_smiles("c1ccccc1[N+](=O)[O-]")
result = nitrobenzene.evaluate_hazard()
print(f"Nitrobenzene: {result.hazard_class}")  # MEDIUM

LOW Hazard¶

# Ethanol
ethanol = Compound.from_smiles("CCO")
result = ethanol.evaluate_hazard()
print(f"Ethanol: {result.hazard_class}")  # LOW

Interpreting Results¶

High Hazard Compounds¶

Immediate concern: High decomposition energy
Handling: Extreme caution, specialized facilities
Validation: Essential experimental characterization

Medium Hazard Compounds¶

Potential concern: One or more criteria triggered
Handling: Standard chemical safety procedures
Validation: Recommended for new compounds

Low Hazard Compounds¶

General chemicals: No CHETAH criteria triggered
Handling: Normal laboratory safety
Note: Still follow proper chemical safety

Decomposition Products¶

Access the predicted decomposition products:

result = compound.evaluate_hazard()

print("Decomposition products:")
for product, moles in result.product_breakdown.items():
    if moles > 0.01:
        print(f"  {product}: {moles:.2f} mol")

Gas Generation¶

Predict gas generation from decomposition:

result = compound.evaluate_hazard()

print(f"Gas volume: {result.gas_volume_L_g:.4f} L/g at STP")

This is useful for:

Containment requirements
Pressure buildup assessment
Ventilation planning

Working with Multiple Compounds¶

Manual Loop¶

smiles_list = [
    "CCO",                                    # Ethanol
    "c1ccccc1[N+](=O)[O-]",                   # Nitrobenzene
    "Cc1c([N+](=O)[O-])cc([N+](=O)[O-])cc1[N+](=O)[O-]",  # TNT
]

for smiles in smiles_list:
    compound = Compound.from_smiles(smiles)
    result = compound.evaluate_hazard()
    print(f"{result.formula}: {result.hazard_class}")

Using Batch Screening¶

For many compounds, use batch processing:

from phoenix import screen

results = screen(smiles_list)
df = results.dataframe

# Filter by hazard class
high_hazard = df[df['hazard_class'] == 'HIGH']
print(f"High hazard compounds: {len(high_hazard)}")

See Batch Processing for more details.

Integration with Other Analysis¶

Combining with Decomposition Analysis¶

compound = Compound.from_smiles("c1ccccc1[N+](=O)[O-]")

# Quick hazard check
result = compound.evaluate_hazard()

# Detailed decomposition analysis
if result.hazard_class != "LOW":
    decomp = compound.max_decomposition(method="both")
    print(f"Hierarchy: {decomp.hierarchy_result.delta_hd_cal_g:.1f} cal/g")
    print(f"LP: {decomp.lp_result.delta_hd_cal_g:.1f} cal/g")
    print(f"Deviation: {decomp.deviation_percent:.1f}%")

Functional Group Details¶

from phoenix.hazard.functional_groups import detect_functional_groups

compound = Compound.from_smiles("c1ccccc1[N+](=O)[O-]")
alerts = detect_functional_groups(compound)

for alert in alerts:
    print(f"{alert.name} (x{alert.count})")
    print(f"  SMARTS: {alert.smarts}")
    print(f"  Risk: {alert.description}")

Best Practices¶

Always validate experimentally for new or unfamiliar compounds
Use batch screening for large datasets
Check functional groups for specific hazard mechanisms
Consider context - LOW doesn't mean "safe"
Document results for safety reviews

Next Steps¶

Decomposition: Detailed decomposition analysis
Functional Groups: Reactive group detection
Batch Processing: Screen multiple compounds