Corrosion Engineering MCP Server

""" Tier 2 Tool: Localized Corrosion Calculator (Pitting & Crevice) Calculates pitting and crevice corrosion susceptibility using: - PREN-based CPT (Critical Pitting Temperature) correlations - Chloride threshold models - Oldfield-Sutton IR drop for crevice acidification Inputs: - Material (e.g., "316L", "2205", "254SMO") - Temperature (°C) - Chloride concentration (mg/L, from Phase 1 PHREEQC) - pH (from Phase 1 PHREEQC) - Crevice geometry (optional) Outputs: - Pitting susceptibility (CPT, PREN, Cl⁻ threshold) - Crevice susceptibility (CCT, IR drop, acidification) - Overall risk assessment - Recommendations Performance: 1-2 seconds (Tier 2 target) Accuracy: ±5°C for CPT, ±20% for Cl⁻ threshold Per Codex guidance: - Expose PREN calibration coefficients (duplex ±5°C deviation) - Separate pitting vs crevice outputs - Share chloride threshold logic - Simplified Oldfield-Sutton IR drop """ from typing import Dict, Optional, Tuple import logging from core.localized_backend import LocalizedBackend, MaterialComposition logger = logging.getLogger(__name__) def _detect_tier_disagreement( tier1_susceptibility: str, tier2_risk: Optional[str], ) -> Tuple[bool, Optional[str]]: """ Detect when Tier 1 (PREN/CPT) and Tier 2 (E_pit vs E_mix) give different risk assessments. Per Codex recommendation: Expose conflicts explicitly with guidance. Args: tier1_susceptibility: "low", "moderate", "high", "critical" tier2_risk: "low", "moderate", "high", "critical", or None (if Tier 2 unavailable) Returns: (disagreement_detected: bool, explanation: str or None) Example: >>> disagreement, msg = _detect_tier_disagreement("critical", "low") >>> print(disagreement) # True >>> print(msg) "Tier 1 (PREN/CPT) says 'critical' but Tier 2 (E_pit vs E_mix) says 'low'. ..." """ if tier2_risk is None: return (False, None) # No disagreement if Tier 2 unavailable # Risk severity ranking (for comparison) risk_levels = {"low": 0, "moderate": 1, "high": 2, "critical": 3} tier1_level = risk_levels.get(tier1_susceptibility.lower(), -1) tier2_level = risk_levels.get(tier2_risk.lower(), -1) # Disagreement = difference of 2+ levels (e.g., "critical" vs "low") if abs(tier1_level - tier2_level) >= 2: explanation = ( f"⚠️ TIER DISAGREEMENT: Tier 1 (PREN/CPT empirical) says '{tier1_susceptibility}' " f"but Tier 2 (E_pit vs E_mix mechanistic) says '{tier2_risk}'. " f"Recommendation: Trust Tier 2 for accurate assessment - it accounts for actual " f"electrochemical driving force and redox conditions. Tier 1 CPT is a conservative " f"screening tool (worst-case ferric chloride test, does not account for dissolved oxygen)." ) return (True, explanation) return (False, None) def calculate_localized_corrosion( material: str, temperature_C: float, Cl_mg_L: float, pH: float = 7.0, crevice_gap_mm: float = 0.1, water_chemistry_json: Optional[str] = None, dissolved_oxygen_mg_L: Optional[float] = None, ) -> Dict: """ Calculate pitting and crevice corrosion susceptibility. Args: material: Material name (e.g., "316L", "2205", "254SMO") temperature_C: Operating temperature (°C) Cl_mg_L: Chloride concentration (mg/L) pH: Solution pH (default 7.0) crevice_gap_mm: Crevice gap width in mm (default 0.1 mm) water_chemistry_json: Optional JSON with full water chemistry for PHREEQC integration dissolved_oxygen_mg_L: Dissolved oxygen concentration (mg/L). If provided, enables Tier 2 electrochemical pitting assessment (E_pit vs E_mix) for NRL materials (HY80, HY100, SS316). Tier 1 PREN/CPT always calculated. Returns: Dictionary containing: - pitting: Pitting susceptibility results (Tier 1 + optional Tier 2) Tier 1 (PREN/CPT - always present): - CPT_C: Critical Pitting Temperature (°C) - PREN: Pitting Resistance Equivalent Number - Cl_threshold_mg_L: Chloride threshold (mg/L) - susceptibility: "low", "moderate", "high", "critical" - margin_C: Temperature margin to CPT (°C) - interpretation: Text summary Tier 2 (Electrochemical - if DO provided and NRL material): - E_pit_VSCE: Pitting initiation potential (V vs SCE), None if not calculated - E_mix_VSCE: Mixed/corrosion potential (V vs SCE), None if not calculated - electrochemical_margin_V: ΔE = E_mix - E_pit (V), None if not calculated - electrochemical_risk: "critical", "high", "moderate", "low", None if not calculated - electrochemical_interpretation: Text summary for Tier 2, None if not calculated - crevice: Crevice susceptibility results - CCT_C: Critical Crevice Temperature (°C) - IR_drop_V: IR drop in crevice (V) - acidification_factor: pH drop factor - susceptibility: "low", "moderate", "high", "critical" - margin_C: Temperature margin to CCT (°C) - interpretation: Text summary - overall_risk: "low", "moderate", "high", "critical" - recommendations: List of mitigation strategies Example: >>> result = calculate_localized_corrosion( ... material="316L", ... temperature_C=60.0, ... Cl_mg_L=500.0, ... pH=6.5, ... crevice_gap_mm=0.1 ... ) >>> print(result["pitting"]["CPT_C"]) 14.5 >>> print(result["overall_risk"]) "high" Material Guidelines: - 304: PREN ≈ 18, CPT ≈ 8°C (low resistance) - 316/316L: PREN ≈ 24, CPT ≈ 14°C (moderate resistance) - 2205 (duplex): PREN ≈ 35, CPT ≈ 25°C (good resistance) - 254SMO: PREN ≈ 43, CPT ≈ 38°C (excellent resistance) Interpretation Guide: - CPT margin > 20°C: Low risk - CPT margin 10-20°C: Moderate risk, monitor chlorides - CPT margin 0-10°C: High risk, mitigation required - CPT margin < 0°C: Critical risk, immediate action required Chloride Thresholds (20°C baseline): - 304: ~50 mg/L - 316L: ~250 mg/L - 2205: ~1000 mg/L - 254SMO: ~5000 mg/L (Decreases exponentially with temperature) Raises: ValueError: If invalid material or parameters """ # Validate inputs if temperature_C < 0 or temperature_C > 150: raise ValueError(f"Temperature {temperature_C}°C out of range (0-150°C)") if Cl_mg_L < 0: raise ValueError(f"Chloride concentration {Cl_mg_L} mg/L must be positive") if pH < 0 or pH > 14: raise ValueError(f"pH {pH} out of range (0-14)") if crevice_gap_mm <= 0 or crevice_gap_mm > 10: raise ValueError(f"Crevice gap {crevice_gap_mm} mm out of range (0-10 mm)") # Run localized corrosion calculation backend = LocalizedBackend() result = backend.calculate_localized_corrosion( material=material, temperature_C=temperature_C, Cl_mg_L=Cl_mg_L, pH=pH, crevice_gap_mm=crevice_gap_mm, dissolved_oxygen_mg_L=dissolved_oxygen_mg_L, ) # Format output output = { "pitting": { # Tier 1 (PREN/CPT - always present) "CPT_C": round(result.pitting.CPT_C, 1), "PREN": round(result.pitting.PREN, 1), "Cl_threshold_mg_L": round(result.pitting.Cl_threshold_mg_L, 1), "susceptibility": result.pitting.susceptibility, "margin_C": round(result.pitting.margin_C, 1), "interpretation": result.pitting.interpretation, # Tier 2 (Electrochemical - optional, requires DO and NRL material) "E_pit_VSCE": round(result.pitting.E_pit_VSCE, 3) if result.pitting.E_pit_VSCE is not None else None, "E_mix_VSCE": round(result.pitting.E_mix_VSCE, 3) if result.pitting.E_mix_VSCE is not None else None, "electrochemical_margin_V": round(result.pitting.electrochemical_margin_V, 3) if result.pitting.electrochemical_margin_V is not None else None, "electrochemical_risk": result.pitting.electrochemical_risk, "electrochemical_interpretation": result.pitting.electrochemical_interpretation, }, "crevice": { "CCT_C": round(result.crevice.CCT_C, 1), "IR_drop_V": round(result.crevice.IR_drop_V, 4), "acidification_factor": round(result.crevice.acidification_factor, 1), "susceptibility": result.crevice.susceptibility, "margin_C": round(result.crevice.margin_C, 1), "interpretation": result.crevice.interpretation, }, "material": result.material, "temperature_C": temperature_C, "Cl_mg_L": Cl_mg_L, "pH": pH, "overall_risk": result.overall_risk, "recommendations": [], } # Generate recommendations recommendations = [] # Overall risk recommendations if result.overall_risk == "critical": recommendations.append( "CRITICAL: Immediate risk of localized corrosion - Material change or process modification required" ) recommendations.append( "Consider upgrading to higher PREN alloy (e.g., 316→2205, 2205→254SMO)" ) elif result.overall_risk == "high": recommendations.append( "HIGH RISK: Localized corrosion likely - Implement mitigation measures" ) recommendations.append( "Options: Reduce temperature, lower chlorides, or upgrade material" ) elif result.overall_risk == "moderate": recommendations.append( "MODERATE RISK: Monitor for pitting/crevice initiation" ) recommendations.append( "Establish inspection schedule (quarterly recommended)" ) else: recommendations.append( "LOW RISK: Material selection appropriate for operating conditions" ) # Pitting-specific recommendations if result.pitting.margin_C < 10.0: recommendations.append( f"Pitting risk: T = {temperature_C}°C near CPT = {result.pitting.CPT_C:.1f}°C" ) if Cl_mg_L > result.pitting.Cl_threshold_mg_L: recommendations.append( f"Reduce chlorides from {Cl_mg_L:.0f} to <{result.pitting.Cl_threshold_mg_L:.0f} mg/L" ) # Crevice-specific recommendations if result.crevice.susceptibility in ["high", "critical"]: recommendations.append( "Crevice corrosion risk: Eliminate or seal crevices in design" ) recommendations.append( "Avoid gasketed joints, threaded connections, or lap joints where possible" ) if result.crevice.acidification_factor > 100: recommendations.append( f"Severe crevice acidification (pH drop factor {result.crevice.acidification_factor:.0f}x) - Use solid sections or welded construction" ) # pH recommendations if pH < 6.0: recommendations.append( f"Low pH ({pH:.1f}) increases localized corrosion risk - Consider pH control >6.5" ) # Material-specific recommendations if "304" in material: recommendations.append( "304 has low pitting resistance (PREN ≈ 18) - Upgrade to 316L (PREN ≈ 24) if chlorides >100 mg/L" ) if "316" in material and Cl_mg_L > 500: recommendations.append( "316 moderate for high chlorides - Consider duplex 2205 (PREN ≈ 35) for Cl⁻ >500 mg/L" ) # Temperature control recommendations if result.pitting.margin_C < 5.0 and result.pitting.margin_C > 0: recommendations.append( f"Reduce operating temperature by {5.0 - result.pitting.margin_C:.1f}°C to gain 5°C safety margin" ) output["recommendations"] = recommendations # Add note about water chemistry integration (future) if water_chemistry_json: output["note"] = "Water chemistry integration with PHREEQC pending (Phase 2 enhancement)" # Detect Tier 1 vs Tier 2 disagreement (per Codex recommendation) disagreement_detected, disagreement_msg = _detect_tier_disagreement( tier1_susceptibility=result.pitting.susceptibility, tier2_risk=result.pitting.electrochemical_risk, ) if disagreement_detected: output["tier_disagreement"] = { "detected": True, "tier1_assessment": result.pitting.susceptibility, "tier2_assessment": result.pitting.electrochemical_risk, "explanation": disagreement_msg, } # Also prepend to recommendations for visibility recommendations.insert(0, disagreement_msg) else: output["tier_disagreement"] = { "detected": False, } return output def calculate_pren( Cr_wt_pct: float, Mo_wt_pct: float, N_wt_pct: float, grade_type: str = "austenitic", ) -> Dict: """ Calculate PREN (Pitting Resistance Equivalent Number) from composition. Utility function for custom alloys or verification. Args: Cr_wt_pct: Chromium content (wt%) Mo_wt_pct: Molybdenum content (wt%) N_wt_pct: Nitrogen content (wt%) grade_type: "austenitic", "duplex", or "superaustenitic" Returns: Dictionary with: - PREN: Calculated value - CPT_C: Estimated Critical Pitting Temperature (°C) - grade_type: Input grade type Example: >>> result = calculate_pren(Cr_wt_pct=16.5, Mo_wt_pct=2.0, N_wt_pct=0.05) >>> print(result["PREN"]) 23.3 >>> print(result["CPT_C"]) 13.3 Standard PREN formula (austenitic): PREN = %Cr + 3.3×%Mo + 16×%N Duplex formula: PREN = %Cr + 3.3×%Mo + 30×%N (higher N weighting) CPT correlation (austenitic): CPT ≈ PREN - 10 (°C) """ # Validate inputs if Cr_wt_pct < 0 or Cr_wt_pct > 30: raise ValueError(f"Cr content {Cr_wt_pct}% out of range (0-30%)") if Mo_wt_pct < 0 or Mo_wt_pct > 10: raise ValueError(f"Mo content {Mo_wt_pct}% out of range (0-10%)") if N_wt_pct < 0 or N_wt_pct > 1: raise ValueError(f"N content {N_wt_pct}% out of range (0-1%)") # Create material composition comp = MaterialComposition( Cr=Cr_wt_pct, Mo=Mo_wt_pct, N=N_wt_pct, grade_type=grade_type, ) # Calculate PREN pren = comp.calculate_pren() # Estimate CPT from core.localized_backend import CPT_CORRELATIONS cpt_corr = CPT_CORRELATIONS.get(grade_type, CPT_CORRELATIONS["austenitic"]) cpt = cpt_corr["m"] * pren + cpt_corr["b"] return { "PREN": round(pren, 1), "CPT_C": round(cpt, 1), "grade_type": grade_type, "composition": { "Cr_wt_pct": Cr_wt_pct, "Mo_wt_pct": Mo_wt_pct, "N_wt_pct": N_wt_pct, }, }