Magnesium chloride does attack concrete and rebar, but the damage is conditional rather than automatic. Chloride ions drive rebar corrosion once they exceed a critical threshold at the steel surface, and magnesium ions can degrade the cement paste itself. Dry, sealed, or non-reinforced concrete often performs acceptably for years, while wet, reinforced, or cracked concrete deteriorates faster than most owners expect.
The short answer hides the useful part. Whether your slab, foundation wall, or garage floor is at risk depends on which damage mechanism is active, how much free water carries the salt into the pore structure, and whether steel sits close enough to the surface to feel it. This explainer separates the mechanisms, quantifies the thresholds, and gives you decision points you can actually check on site.
Two mechanisms behind magnesium chloride damage
Two distinct processes run in parallel, and they operate at different depths and timescales.
The first is chloride-induced corrosion of reinforcing steel. Chloride ions penetrate the concrete's pore network and accumulate at the steel surface. There they break down the passive iron-oxide film that normally protects rebar in alkaline concrete. Once that film is compromised, oxygen and water complete a corrosion cell, and the rust that forms occupies roughly two to six times the volume of the original steel. That expansion generates tensile stress inside the concrete, which cracks it from the inside out. Cracking then admits more chloride and water, so the process accelerates.
The second is magnesium attack on the cement hydration products. Concrete strength comes largely from calcium silicate hydrate, often abbreviated C-S-H, the glue that binds aggregate together. Magnesium ions react with C-S-H and with calcium hydroxide, converting them into magnesium silicate hydrate and brucite, a softer and less cohesive phase. This is sometimes called magnesium sulfate attack when sulfates are present, but the magnesium cation does its own work. The result is a reduced-binding, more porous paste near the surface, which then sheds aggregate and scales.
The plain-language version: chloride goes after the steel, magnesium goes after the glue, and water is the delivery vehicle for both. Remove the water and you slow both mechanisms dramatically.
How the salt gets in matters too. Magnesium chloride is strongly hygroscopic, meaning it pulls moisture from the air and holds it. That property is why it works as a dust suppressant and deicer, and it is also why treated concrete stays wetter longer than untreated concrete. A surface that stays damp keeps feeding both mechanisms.
Quantified context: thresholds, concentrations, and exposure
The numbers that matter most are chloride thresholds at the steel, not chloride concentrations in the bag or on the surface.
A widely cited threshold for chloride-induced corrosion in reinforced concrete is roughly 0.05 percent chloride by mass of cement for prestressed concrete, and around 0.10 to 0.15 percent by mass of cement for conventional reinforced concrete exposed to chlorides. These are order-of-magnitude guides rather than universal pass-fail lines; the actual threshold shifts with pH, oxygen availability, moisture, and the quality of the concrete. Some standards use a simpler whole-concrete basis, commonly around 0.03 to 0.05 percent chloride by mass of concrete for reinforced members in wet service. The two bases are not interchangeable, which is a common source of confusion in field reports.
Reconcile them this way. Cement-based thresholds are higher numbers because cement is only about 10 to 15 percent of a concrete mix by mass. A 0.10 percent by-cement threshold is roughly 0.010 to 0.015 percent by-concrete, not 0.10 percent by-concrete. If a lab report gives you one basis and a specification gives you the other, convert before comparing. Comparing raw numbers across bases is the single most common error in chloride assessments.
Dose and concentration matter as well. Magnesium chloride deicers are typically sold as flakes or as liquid solutions in the 25 to 30 percent concentration range. Dilution by snowmelt, rain, and runoff lowers the surface concentration, but repeated application can keep the surface pore solution above the corrosion threshold for extended periods.
Depth of cover is a practical variable you can measure. Rebar with 2 inches of good-quality, low-permeability cover resists chloride ingress far longer than rebar with 0.75 inch of cover. That is why garage slabs, parking structures, and bridge decks specify cover depths and why shallow cover is a red flag regardless of what salt is used.
Comparison of chloride deicers and their concrete effects
Not all chloride salts behave the same way, and the differences are large enough to change a specification.
| Deicer | Primary concrete concern | Relative chloride load | Notes |
|---|---|---|---|
| Sodium chloride (rock salt) | Chloride corrosion of rebar | High | Cheapest; chloride-driven corrosion dominates |
| Calcium chloride | Chloride corrosion; exothermic and aggressive | High | Fast melting; strong hygroscopic pull |
| Magnesium chloride | Chloride corrosion plus magnesium paste attack | High | Hygroscopic; also used as dust suppressant |
| Calcium magnesium acetate | Low chloride; acetate can still attack some binders | Very low | Costlier; marketed as lower-corrosion |
| Potassium acetate | Low chloride; acetate-related concerns in some mixes | Very low | Common at airports |
| Sand or grit | None chemically | None | No melting action; traction only |
The counterintuitive detail: switching from magnesium chloride to another chloride deicer does not solve a corrosion problem. Sodium chloride and calcium chloride deliver chloride just as readily. If rebar corrosion is the concern, the meaningful comparison is chloride versus non-chloride, not one chloride versus another.
Timeline of deterioration in reinforced concrete
Damage from magnesium chloride is not instant. It follows a recognizable sequence, and the timescales below are typical ranges for reinforced concrete in wet, chloride-exposed service, not guarantees for a specific structure.
| Stage | What happens | Typical timeframe |
|---|---|---|
| Surface wetting | Salt solution saturates surface pores | Minutes to hours per event |
| Chloride ingress | Chloride migrates inward through pore water | Months to years, depending on cover and permeability |
| Threshold reached at steel | Chloride exceeds critical level at rebar | Often 5 to 15 years at shallow cover; longer at deep cover |
| Initiation of corrosion | Passive film breaks down locally | Weeks to months after threshold |
| Active corrosion and cracking | Rust expansion cracks cover concrete | 1 to 5 years after initiation |
| Delamination and spalling | Cracks connect; concrete separates | Years, accelerating with continued salt exposure |
Read the table as a sequence, not a schedule. A well-built slab with deep cover and low permeability can sit in the ingress stage for decades. A cracked, shallow-cover slab can move through initiation in a few seasons. Freeze-thaw cycling compresses every stage, because each cycle pumps solution deeper and opens new microcracking.
Edge cases and where the risk concentrates
Several conditions change the answer materially.
Non-reinforced concrete has no steel to corrode, so the dominant risk becomes magnesium attack and surface scaling. Plain concrete flatwork exposed to wet magnesium chloride can still pit and scale at the surface over years, but it will not suffer rebar-driven delamination.
Dry exposure is much lower risk. Magnesium chloride applied to a sealed, well-drained, covered slab that dries between events has limited opportunity to move chloride to the steel. Persistent wetting is the amplifier.
Freeze-thaw exposure is the worst combination. Concrete that cycles between wet and frozen states while carrying chloride experiences both salt attack and physical frost damage. Air-entrained concrete resists the frost component better than non-air-entrained concrete, but air entrainment does not stop chloride ingress.
Cracks and joints are preferred pathways. A 0.012-inch crack is enough to admit solution and move chloride past the cover. Joints, cold pours, and construction defects concentrate risk, which is why inspection should start there.
Precast, prestressed, and post-tensioned elements deserve extra caution. Prestressed steel is more sensitive to chloride-induced corrosion and can fail in a brittle manner, so many specifications set lower chloride limits for prestressed concrete. If you manage a parking structure with post-tensioned tendons, treat any chloride exposure as high-priority.
Misconceptions worth correcting
One persistent misconception is that magnesium chloride is safe because it is a "salt" and salts are natural. The word natural carries no corrosion meaning. Chloride is chloride wherever it comes from, and magnesium is a real cementitious aggressor.
A second misconception is that a sealer makes the concrete immune. Sealers reduce ingress rates and can buy years, but they degrade under traffic and freeze-thaw cycling. A sealer is a control measure, not a barrier that eliminates risk.
A third is that staining, whitening, or surface powdering is cosmetic. Some of it is cosmetic, but efflorescence and surface softening near joints can indicate paste degradation from magnesium attack. If surface material rubs away easily, investigate rather than repaint.
The professional caveat worth stating plainly: most "does salt damage concrete" claims in the field are not based on measured chloride content at the steel. They are based on visible surface damage. Surface appearance and internal corrosion progress are poorly correlated, especially in early stages. A slab can look acceptable at the surface while chloride levels at the steel are already climbing toward threshold. Do not let a clean surface end the investigation.
Practical actions and decision thresholds
You can manage magnesium chloride risk with a short, concrete set of steps.
Test before you conclude. Collect powder samples at two or three depths, typically a shallow sample and a sample near the cover depth above the rebar, and test for acid-soluble chloride. Compare results on a consistent basis: by mass of cement or by mass of concrete, matched to your specification. Know your thresholds. Treat 0.05 percent chloride by mass of cement as an alert level for prestressed members and around 0.10 to 0.15 percent by mass of cement as the concern zone for conventional rebar. Convert to a concrete basis if your lab reports that way. Measure cover depth. A cover meter over a few representative areas tells you whether rebar sits at 0.75 inch or at 2 inches. Shallow cover plus any chloride exposure is a high-priority condition. Control the water. Drain or slope surfaces so solution does not pond, and avoid applying chloride deicer where it drains toward foundations, joints, or slab edges. Inspect the usual suspects first. Look along joints, cracks, and edges for rust staining, delamination, and soft surface paste. Tap-test suspect areas for hollow-sounding concrete. Use non-chloride deicers where corrosion risk is high. Calcium magnesium acetate and potassium acetate reduce chloride load, at higher cost per application. Reserve them for structures where rebar corrosion is the dominant concern.
Red-flag warnings. Any of the following should trigger a professional evaluation rather than a wait-and-see approach.
Rust staining or brown bleed marks near joints or cracks on a slab that receives magnesium chloride. Any structure with prestressed or post-tensioned steel and confirmed chloride exposure. Chloride at the steel that meets or exceeds the applicable limit on the same basis as the specification. Hollow-sounding cover concrete, indicating delamination from active corrosion. Ponding water that persists for days after applications, keeping the surface continuously wet. Repeated deicer use on the same slab with no drainage and no sealer maintenance.
Limitations of what the evidence supports
Two limits deserve emphasis. First, chloride thresholds are site-specific and basis-dependent, so use them as screening tools and confirm with a qualified materials engineer before making structural decisions. Second, this explainer addresses mechanisms and general exposure conditions. It does not substitute for mixture design review, cover inspection, or a corrosion survey on a specific structure.
The honest bottom line is that magnesium chloride is a real concrete and rebar hazard whose severity is set by moisture, chloride concentration, cover depth, and concrete quality. You cannot eliminate risk by choosing a different salt within the chloride family. You can control it with drainage, cover, sealer maintenance, and test-based decision points that tell you when surface appearance has stopped being a trustworthy indicator.