Thermogravimetric Analysis of Flux Solvent Evaporation in Reflow Profiles

Mass

Precision thermal gravimetry measures weight loss in a solder paste deposit as furnace temperatures rise under a controlled atmosphere. Solvent vehicles make up eight to fifteen percent of total paste weight, which translates to roughly fifty percent of its liquid volume. Thermogravimetric analysis tracks the evaporation windows of these volatile organic compounds at heating rates that match conveyorized reflow ovens.

When a ten-milligram sample is heated, the resulting mass-loss curve plots weight against temperature and time to show how each solvent fraction volatilizes.

Solder paste fluxes use multi-component solvent packages to balance stencil printing rheology against controlled outgassing in the reflow oven. Low-boiling fractions evaporate between 60°C and 130°C to manage tack and viscosity during storage and printing. High-boiling solvents remain liquid through the initial ramp, carrying active chemicals past 170°C. If these high-boiling fractions boil off too early in the ramp phase, solid activators drop out of solution and halt oxide removal before the alloy melts.

Evaluating these mass-loss profiles requires high-resolution balances operating in a controlled nitrogen or air stream. Microgram sensitivity captures small changes in slope, giving engineers the data needed to calculate evaporation activation energies through dynamic or isothermal methods. Dynamic runs apply constant heating rates ~ 0.5°C, 1.0°C, or 2.0°C per second ~ from ambient up to 300°C. Isothermal testing holds the sample at a single plateau like 150°C, tracking weight loss over time to distinguish solvent evaporation from activator breakdown.

Dynamic Temperature Ramp Thermogravimetry

Thermogravimetric curves plotted at constant ramp rates reveal distinct inflection points corresponding to boiling thresholds and heats of vaporization. Ramp rates directly shift where these evaporation windows appear on the temperature scale, since faster heating causes internal sample temperatures to lag behind furnace thermocouples. Testing a flux at 0.5°C per second shows complete solvent loss by 145°C, whereas ramping at 2.0°C per second leaves residual solvent in the sample up to 175°C.

Analyzing thermogravimetric curves pinpoints exact solvent loss thresholds before the solder alloy reaches liquidus. Taking the derivative of mass against temperature or time produces a differential curve that separates individual peaks for each solvent component. A dual-solvent flux generates two distinct peaks: a primary peak at 110°C for the low-boiling ether and a secondary peak at 165°C for the high-boiling glycol.

Mapping these peaks against reflow zone profiles ensures the oven drives off the primary solvent while retaining enough secondary carrier to maintain flux viscosity through the soak.

Solvent Vapor Pressure and Evaporation Rates

Evaporation rates of organic solvents in printed paste deposits follow modified Langmuir equations, where mass loss per unit area depends on vapor pressure, molecular weight, and gas velocity. High-boiling solvents like diethylene glycol dibutyl ether or dipropylene glycol monobutyl ether have low room-temperature vapor pressures, stabilizing stencil life on the line. Once the assembly enters the preheat zone, rising temperatures boost solvent vapor pressure exponentially while convective airflow sweeps away the stagnant boundary layer above the deposit, accelerating evaporation beyond what static lab tests indicate.

The evaporation rate of high-boiling glycol ether solvents inside a forced-convection reflow zone accelerates by up to three times compared to static nitrogen thermogravimetry balances operating at identical temperature setpoints.

Formulations balance the partial pressures of constituent solvents so light fractions do not boil off violently. Rapid boiling traps gas pockets in the paste deposit, displacing metal powder and spattering solder onto the mask. If low-boiling solvents make up more than twenty percent of total solvent mass, expanding gas triggers micro-bursts during preheat.

Mass loss rates exceeding 0.15 percent weight loss per degree Celsius signal aggressive outgassing, requiring a longer thermal soak to preserve deposit integrity.

Carrying residual solvents past the alloy melting point creates serious reliability issues in finished joints. Trapped liquid decomposes at peak temperatures, venting gas into molten solder and forming voids beneath bottom-terminated components. Thermogravimetric data shows that retaining more than 0.5 weight percent solvent at alloy liquidus pushes voiding past IPC-A-610 Class 3 limits.

Selecting a flux vehicle whose mass-loss curve drops to baseline before melting eliminates solvent-induced voiding at the source.

Short thermal soak times leave unevaporated high-boiling solvents trapped beneath low-clearance components. When these solvents boil during peak reflow, they splatter flux onto adjacent fine-pitch traces, causing shorts.

Volatiles

Volatile compounds in solder flux encompass primary solvents, co-solvents, rheology modifiers, and volatile activators. Thermogravimetric testing separates physical solvent evaporation from outgassing caused by chemical reactions. Physical evaporation occurs at lower temperatures without changing molecular structure.

Chemical outgassing follows as organic activators, such as dicarboxylic acids, reduce metal oxides on component leads and board pads, releasing water vapor, carbon dioxide, and metal-organic salts.

Evaluating flux formulations involves comparing differential mass derivative peaks directly against defect counts from automated X-ray inspection. Differential thermogravimetry plots the rate of mass change in milligrams per minute or percentage weight loss per degree Celsius. When derivative peaks extend past the liquidus temperature of lead-free alloys like SAC305 (217°C), preheating was incomplete or the vehicle lacks thermal stability.

Liquids trapped beneath molten metal form vapor bubbles that cannot escape before the solder solidifies, leaving large voids in thermal pad joints.

Differential Derivative Mass Loss Interpretation

Interpreting differential mass-loss curves requires pinpointing the thermal onset, peak rate temperature, and baseline return for each volatile component. Thermal onset marks the point where weight loss rises above balance noise, typically 0.05 percent of sample weight. Peak rate temperature corresponds to maximum evaporation velocity, where solvent vapor pressure balances convective gas flow.

Baseline return confirms complete removal of that volatile fraction, leaving non-volatile resins, rosin matrices, and active fluxing agents on the board.

Single-stage solvent systems produce a sharp, narrow derivative curve, reflecting rapid volatilization over a window of fifteen to twenty degrees Celsius. Multi-stage systems generate broader profiles with overlapping peaks, spreading solvent release across forty to eighty degrees. This extended release widens the process window on heavy assemblies where varying component masses heat unevenly.

Matching the solvent profile to the board’s thermal delta prevents early dry-out on small components while ensuring full solvent removal from heavier ones.

Carrier Vehicle Solvent Formulations

Carrier solvent selection dictates both printing behavior and required reflow profiles. Formulators evaluate boiling points, flash points, polarities, and resin solubility when picking solvents. Common choices include aliphatic alcohol ethers, ester alcohols, and polyhydric glycol derivatives.

Each family exhibits distinct mass-loss kinetics during dynamic heating, determining how the paste deposit responds to shifts in conveyor speed and zone temperatures.

High-boiling solvents like tetraethylene glycol dimethyl ether offer long stencil life, maintaining stable print performance across eight-hour shifts. If the preheat phase is too brief, however, their high boiling points leave elevated residual mass at the 217°C liquidus mark. Lower-boiling options like hexylene glycol clear cleanly during short preheat cycles, but risk drying on the stencil during line pauses.

Qualification balances print stability against outgassing through empirical thermogravimetric screening.

• Premature Solvent Dryout occurs when low-boiling fractions evaporate on the stencil or in early preheat zones, leading to paste stiffening, incomplete aperture release, and poor wet-out on oxidized leads.
• Solvent Entraptment Voids happen when residual high-boiling solvents stay in the paste past alloy liquidus, expanding into gas bubbles inside molten solder joints under QFN thermal pads.
• Solder Spatter Release occurs when boiling solvent trapped beneath viscous rosin builds pressure until bursting through the surface, scattering solder spheres across the solder mask.
• Flux Slump Displacement occurs when slow-evaporating solvents thin the resin matrix too much between 120°C and 150°C, causing printed deposits to bridge across fine-pitch pads.

Screening solvent outgassing shows how individual formulations behave under rapid ramp rates. Elevated voiding on bottom-terminated components can stem from stencil design or reflow settings rather than solvent evaporation kinetics alone.

Decomposition

Beyond physical evaporation, flux vehicles undergo pyrolytic decomposition at higher temperatures. Thermal gravimetry tracks how solid rosin binders, synthetic resins, and carboxylic acid activators break down near peak reflow setpoints. Rosin matrices ~ primarily abietic and pimaric acid isomers ~ soften at 70°C, liquefy by 120°C, and decompose between 230°C and 280°C. Knowing where solvent release ends and resin breakdown begins ensures adequate fluxing without leaving charred, un-cleansable residue.

Organic acid activators remove oxalates, oxides, and hydroxides from copper pads and tin-plated leads through reduction reactions. Common activators like glutaric, succinic, adipic, and sebacic acid each exhibit distinct melting points and stability limits. Succinic acid melts at 184°C and decomposes rapidly above 220°C into volatile anhydrides and water vapor.

If an activator breaks down before the alloy reaches liquidus, metal surfaces re-oxidize in the furnace, leading to non-wetting, de-wetting, or cold joints.

Activator Matrix Thermal Breakdown Mechanics

The thermal breakdown of organic acid activators follows first-order kinetics, where mass-loss rates depend directly on temperature and exposure time. Thermograms of pure activators show steep weight loss across narrow temperature spans. Compounding activators into rosin carriers shifts their decomposition thresholds higher, because the hydrophobic resin matrix encapsulates the active molecules.

Malonic acid breaks down early, making it suitable for low-temperature tin-bismuth alloys that melt at 138°C. Using malonic acid in standard SAC305 profiles destroys the activator long before reaching the 217°C liquidus mark. Conversely, sebacic acid remains stable up to peak temperatures of 260°C, sustaining oxide reduction but leaving higher non-volatile residues that require thorough cleaning in high-reliability aerospace assemblies.

Rosin Rheology and Viscosity Transitions

Rosin and synthetic resin binders keep activator chemicals isolated from solder powder during storage. As furnace heating begins, the rosin melts into a fluid liquid that dissolves activators and coats the alloy particles. Viscosity drops rapidly between 100°C and 160°C, allowing the flux to spread across pads and wet copper surfaces before the alloy melts.

Extended soak times at elevated temperatures exhaust activator reserves, leading to surface oxidation and poor fillet formation at peak reflow. Resin mass loss accelerates above 220°C as rosins undergo decarboxylation and dimerization, releasing heavy hydrocarbon vapors and forming polymerized residue. Retaining sufficient resin mass through peak reflow provides a liquid barrier that shields molten tin-copper intermetallics from atmospheric oxygen.

Rosin thermal decomposition exceeding sixty percent by weight before solder solidification deprives the molten joint of protective barrier coverage, causing immediate surface re-oxidation and severe de-wetting.

Measuring residual mass through post-reflow thermogravimetry indicates how easily residues will clean off in water-wash or no-clean operations. Water-washable chemistries use hydrophilic glycol and amine carriers that decompose into water-soluble compounds. No-clean systems rely on modified hydrogenated rosins that form hard, non-conductive glass layers over unreacted activators.

If thermogravimetric analysis reveals incomplete volatilization of corrosive hydrophilic solvents in a no-clean flux, it signals a risk of electrochemical migration in service.

• Thermal Stability Range establishes the temperature span where organic activators stay chemically intact to reduce metal oxides without undergoing premature pyrolytic breakdown.
• Residue Volatilization Fractional Index measures the percentage of total flux vehicle mass converted to gas before assembly cooling, defining post-reflow cleanliness metrics.
• Decarboxylation Onset Temperature marks the threshold where functional carboxylic acid groups break down into carbon dioxide gas, stripping the flux matrix of its acid cleaning capability.
• Polymerization Residue Threshold identifies the temperature where fluid rosin binders form cross-linked polymeric solids that resist solvent cleaning and aqueous wash cycles.

Flux vehicles designed for lead-free profiles require greater thermal endurance than legacy tin-lead formulations because peak temperatures are higher. Extending thermal exposure beyond the chemical limits of the rosin matrix causes charring, leaving dark, brittle residues that interfere with test probes.

Excessive peak temperatures induce polymer cross-linking in rosin matrices, converting soft, soluble residues into insoluble charred compounds.

Desorption

Desorption kinetics describe how adsorbed solvent molecules and gaseous reaction byproducts release from metal interfaces, substrates, and resin matrices. During reflow, desorption occurs alongside alloy melting and intermetallic formation. As solder powder transitions into a liquid pool, desorbing gases and reaction byproducts must rise through the molten metal column to escape.

Any gas trapped by rapidly freezing solder forms internal voids that degrade electrical, thermal, and mechanical performance.

Voiding under bottom-terminated components, such as QFN thermal pads and power MOSFET drain pads, correlates directly with desorption rates while the alloy remains liquid. Low clearance restricts gas escape, forcing desorbing solvent vapors to migrate horizontally through molten solder. If the volumetric desorption rate exceeds what the standoff clearance can vent, gas pockets coalesce into macro-voids exceeding 25 percent of total pad area.

What Thermal Off-Gassing Rates Indicate Solder Voiding Risk?

Combining thermogravimetric off-gassing measurements with mass spectrometry identifies the chemical species desorbing from flux at liquidus temperatures. Heavy off-gassing of high-boiling solvents, water vapor, or carbon dioxide between 217°C and 245°C signals a high risk of voiding. Specifically, off-gassing rates above 0.02 milligrams per square millimeter per second during time-above-liquidus correlate directly with total joint voiding exceeding fifteen percent.

Coupling a mass spectrometer captures mass-to-charge ratios of desorbed volatiles, distinguishing solvent evaporation from activator degradation. Off-gassing dominated by low-molecular-weight alcohol fragments indicates incomplete preheat evaporation. Conversely, off-gassing rich in carbon dioxide and water vapor points to active oxide reduction inside the molten solder pool, proving flux cleaning reactions occurred too late in the heating cycle.

Outgassing Pressure under Bottom Terminated Components

Desorption beneath large component bodies builds local vapor pressure within the tight gap between die and substrate. This outgassing pressure depends on solvent evaporation rate, flux viscosity, pad area, and standoff height. Elevated pressure pushes molten solder outward beyond pad boundaries, causing micro-solder balls, bridging under component bodies, and irregular joint fillets.

Thermogravimetric mass loss exceeding 0.8 percent of total flux weight during the time-above-liquidus phase increases bottom-terminated component voiding rates beyond acceptable Class 3 thresholds.

Mitigating outgassing pressure requires tuning the soak profile to achieve complete solvent desorption before the alloy reaches liquidus. Thermogravimetric curves define the soak duration needed between 150°C and 180°C to bring residual volatiles below critical thresholds. A longer soak lets volatile components desorb through open paste channels before alloy particles coalesce into an impermeable liquid barrier.

• Volatile Outgassing Rate quantifies the mass of gas desorbed from the flux matrix per unit time during the critical time-above-liquidus window, measured in milligrams per second.
• Venting Channel Standoff Clearance defines the height gap beneath a surface-mount component body that allows desorbing flux gases to escape into the furnace atmosphere.
• Coalescence Volumetric Pressure measures the internal gas pressure generated by expanding solvent vapor pockets trapped within a continuous pool of molten solder alloy.
• Interfacial Desorption Energy calculates the thermal energy needed to release adsorbed solvent molecules from oxidized copper lead-frame surfaces.

Assembly contracts mandate strict compliance with component voiding limits under IPC standards. IPC-7093 guidelines for bottom-terminated component assembly specify a maximum single void area under thermal pads of nine percent, with total combined voiding capped at fifteen to twenty-five percent depending on reliability class.

According to IPC-7093 Section 8.2, thermal pad voiding over fifteen percent total coverage requires engineering review, while single voids exceeding nine percent of the pad surface mandate process modification or board rejection.

Soak

Designing an optimal reflow profile relies on mapping oven heating zones directly to the thermogravimetric evaporation profile of the solder paste. Profiles generally follow one of two formats: Ramp-Soak-Spike or Linear Ramp-to-Spike. Ramp-Soak-Spike uses a dedicated thermal plateau ~ typically 150°C to 180°C for lead-free SAC alloys ~ to equalize temperatures across complex assemblies and drive off volatile solvents.

Linear Ramp-to-Spike profiles maintain a steady temperature rise from ambient to peak, minimizing thermal exposure for heat-sensitive components.

Thermogravimetric data dictates which profile architecture a paste vehicle requires. Formulations with high concentrations of high-boiling solvents demand an extended soak to eliminate volatiles before the alloy melts. Formulations built on rapidly evaporating, low-boiling solvents perform well under linear Ramp-to-Spike profiles, achieving low voiding and high yield while cutting cycle time and energy use.

Ramp Soak Spike Optimization Strategy

Optimizing a Ramp-Soak-Spike profile requires selecting soak entry temperature, exit temperature, and duration. Mass-loss curves identify the temperature range where solvent volatilization peaks. Setting soak entry ten degrees below the mass-loss onset point and exit ten degrees above baseline return ensures complete solvent removal without prematurely depleting active flux compounds.

Across production line audits, increasing soak duration by twenty seconds reduced QFN voiding below twelve percent. Short soaks leave residual solvent in the paste, causing elevated QFN voiding and high solder-ball defect rates from explosive outgassing during the peak phase. Conversely, excessively long high-temperature soaks exhaust active acid fluxing agents prematurely, leading to oxide re-formation, poor wetting, elevated voiding, and solder balling from uncoalesced powder.

Thermal Delta Reduction across High Mass Panels

Complex assemblies with dense copper power planes, heavy heat sinks, and light passives create large thermal mass differentials across the panel. Light components heat quickly, reaching preheat setpoints seconds ahead of heavy ones. A thermal soak allows low-mass areas to plateau while high-mass sections catch up, reducing the thermal delta across the board to under five degrees Celsius before ramping to peak reflow.

Maintaining a thermal delta below five degrees Celsius across panel locations prior to alloy liquidus prevents differential solder melting, eliminating tombstoning on small passive components.

Thermogravimetric validation ensures that the longer exposure needed to equalize heavy panels does not exceed the flux vehicle’s thermal endurance limits. If low-mass zones experience excessive soak times, the solvent vehicle volatilizes completely, causing local dry-out and soldering failures on fine-pitch components. Selecting flux chemistries with broad mass-loss windows provides the process latitude needed to solder heavy panels reliably.

1. Mount calibrated thermocouples to high-mass power inductors, low-mass 0402 passives, and bottom-terminated QFN thermal pads on a test panel.
2. Run the instrumented panel through the reflow oven to capture baseline temperature traces across monitored component locations.
3. Compare board panel thermal traces against the thermogravimetric mass loss curve of the solder paste to identify solvent evaporation windows.
4. Adjust individual zone temperature setpoints and conveyor speed to align the thermal soak plateau with the peak solvent evaporation derivative window.
5. Verify post-reflow joint quality using automated X-ray inspection to confirm total voiding under QFN thermal pads remains below target thresholds.

Balancing soak profile parameters requires trade-offs between component thermal limits, line speed, solvent outgassing, and activator lifetime. Does the thermal soak duration provide sufficient margin for component mass variance without driving the flux chemistry past its thermogravimetric stability limit?

Throughput

Optimizing reflow profiles based on thermogravimetric evaporation data directly impacts assembly line productivity, operational costs, and overall throughput. Oven length, zone count, and conveyor speed dictate cycle time per panel. Extending profile length to accommodate slow-evaporating, high-boiling solvent formulations forces lower conveyor speeds, directly reducing hourly panel output across the line.

A ten-zone convection reflow oven running at 95 centimeters per minute yields a transit time through the heated tunnel of roughly four minutes. Solder paste formulations requiring an extended 120-second soak force operators to reduce conveyor speed to 70 centimeters per minute, cutting line throughput by over twenty-six percent. High-volume manufacturing lines require solder paste vehicles engineered for fast, clean evaporation, supporting rapid linear ramp profiles at conveyor speeds above 110 centimeters per minute.

Conveyor Speed Constraints and Oven Heating Zones

Conveyor speed calculations reconcile heating rates and soak durations against the physical length of the reflow zones. Each heating zone provides independent upper and lower temperature control, allowing process engineers to shape the thermal curve along the conveyor path. Ovens with fewer heating zones, such as six-zone models, struggle to execute complex Ramp-Soak-Spike profiles at high conveyor speeds because steep thermal jumps between short zones induce thermal shock and wide temperature deltas across heavy panels.

High-speed facilities deploy twelve to fourteen-zone reflow furnaces to maintain precise control over solvent volatilization while maximizing panel throughput. Multi-zone configurations enable gradual thermal transitions, establishing gentle evaporation slopes that prevent spatter without slowing the line. Investing in longer ovens transfers process flexibility from the chemical flux vehicle to the mechanical equipment, allowing fast-evaporating paste formulations that maximize line efficiency.

Economic Balance between Cycle Time and Reflow Yield

Calculations balancing line speed against first-pass yields demonstrate that running profiles near the limit of solvent removal margins creates unacceptable commercial risk. A high-speed line producing 120 assemblies per hour generates substantial revenue, but an unoptimized profile that leaves 0.3 percent residual solvent mass increases post-reflow touch-up and scrap costs from voiding escapes and solder balling. SMT line operating costs average between $150 and $350 per hour depending on AOI and X-ray capital depreciation schedules.

Running aggressive, shortened profiles to maximize conveyor speed yields higher nominal panel throughput but carries substantial financial penalties in defect rework and reduced first-pass yield. The optimized profile, built around the thermogravimetric evaporation window of the solvent package, achieves the highest net operational value per hour by balancing high line speed with near-zero outgassing defect rates. Conservative extended soak profiles eliminate outgassing defects entirely, but penalize financial performance through excessive cycle times and lost factory capacity.

Establishing process windows grounded in thermogravimetric mass-loss data ensures that solder paste selection, profile engineering, and conveyor speeds work in harmony. SMT line managers who qualify flux vehicles through quantitative thermal gravimetry eliminate outgassing defects, protect joint integrity, and maximize net financial returns across high-reliability assembly runs.