GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

Hydrophobic Properties of Mineral Wool

woobagoobaa | Posted in General Questions on

Folks.  My understanding from various sources is that mineral wool insulation offers superior fire resistance, good R value, is vapor open, and is hydrophobic.  The latter being particularly important for an upcoming exterior wall dense pack insulation job.  We were preferring dense pack mineral wool over cellulose for the claimed hydrophobic properties.

Inspired by the Wing Nuts, I’ve obtained samples of Termafibre SAFB, Thermafibre Insul FIL, and Rockwool Safe’n’Sound.  All of them getting a  dunk test (see pic).   SAFB and Safe’n’Sound floating top left and right, Insul FIL ball sank like a rock (no pun intended).  Some water does wring out of the SAFB and Safe’n’Sound after being submerged, but I can only describe the Insul FIL as acting like a sponge.  Yet product data sheet claims “ASTM C 1104 Absorbs less than 1% by volume

Hydrophobic?  What gives?

GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.


  1. Expert Member

    Wooga Goobaa,

    Not a direct answer to your question, but the hydrophobic qualities of mineral wool aren't necessarily seen as a positive. Cellulose, being hydroscopic, can buffer and store moisture in the wall during periods of high humidity. This keeps the moisture from causing damage and it can be released when the humidity levels decrease. Builders monitoring thick dense-packed walls, (like the inimitable Dan Kolbert) believe it may be one of the main reasons these walls perform better than predicted.

  2. Expert Member
    BILL WICHERS | | #2

    Mineral wool sinking or floating isn't really the best test here. Mineral wool can dry out, and won't readily absorb moisture. If you want to setup a test, try sticking some mineral wool PARTLY into the water -- water shouldn't wick up the mineral wool.

    BTW, I agree with Malcolm too -- sometimes cellulose' buffering abilities are a positive with moisture control, and mineral wool doesn't offer that particular ability.


    1. woobagoobaa | | #4

      I'll try your test modification. Thanks

  3. woobagoobaa | | #3

    Thank you for your responses. Dense pack cellulose is off the menu for reasons of fire resistance / local requirements. Open/closed cell foam is also off the list. We are looking for max hydrophobic capabilities for the exterior wall dense pack. The exterior wall assembly is (out to in ... zone 5) ... Hardie Plank, 1/4" air gap, Blueskin WRB, 3/4" board sheathing, dense pack insulation, Membrain, skim coated Blueboard, Latex paint. The goal is a water proof, vapor open assembly.

    Forgive me for being dense (ha ha), but why is Insul FiIl sinking and acting like a sponge where the other mineral wool candidates float and appear to absorb far less water?

  4. Expert Member
    NICK KEENAN | | #5

    I'm trying to figure out what you're testing for. Absent extraordinary circumstances, if there is liquid water in your insulation, your wall has failed.

    What you do have to prepare for is high relative humidity, even to the point of condensing. You want insulation that doesn't lose insulation ability under high humidity, and sheds any moisture it absorbed when conditions are right for drying.

  5. Expert Member
    BILL WICHERS | | #6

    >"Dense pack cellulose is off the menu for reasons of fire resistance / local requirements."

    You're in an area where code prohibits dense pack cellulose? I didn't think it was prohibited anywhere. Cellulose insulation is treated with a fire retardent, so it's not really a fire risk any more than the wood structure of a typical house is.

    While mineral wool is very resistant to heat and fire, there are a lot of other things to consider besides just insulation if fire resistance is your goal. Using mineral wool insulation alone won't do a whole lot to improve the fire resistance of your structure.


    1. woobagoobaa | | #7

      >"Dense pack cellulose is off the menu for reasons of fire resistance / local requirements."

      Local requirements == specific requirements of this renovation (not regulatory). Dense urban environment.

  6. Expert Member
    Dana Dorsett | | #8

    Submerging a house or pushing t out onto a lake to see what happens isn't a good way to test it's hydrophobic characteristics. The hydrophobic aspects of mineral are related to trickles of liquid water from intermittent bulk water incursions, not blasting it with a fire hose, not pretending it will float like a raft raft. The bottom line is that it doesn't wick water very strongly- it is the opposite of cellulose in that regard. How strong the wicking forces are will vary by vendor and binding agent.

    ASTM C 1104 is not a liquid water immersion test, but rather the sorption of water vapor a high humidity atmosphere:

    >>""Dense pack cellulose is off the menu for reasons of fire resistance / local requirements."
    >Local requirements == specific requirements of this renovation (not regulatory). Dense urban environment."

    The materials used for siding & sheathing have a much bigger affect on fire resistance from building to building in a dense urban environment than what's in the walls. The high air resistance of cellulose (even at 2lbs density, let alone 3lbs+ dense packed) makes as little as 16" of cellulose in a stud bay qualify as fire blocking(!). While in a full conflagration the fire will eventually cook off the fire retardents in the cellulose, but by the time that actually happens the assembly is already fully engulfed in flame, a total write off.

    An inch of continuous rigid rock wool on the exterior under the siding does a heluva lot more for fire spread building to building than whatever gets stuffed in the wall cavities. Combining that with fiber cement siding and steel shutters over the windows makes the assembly nearly forest-fire proof.

    Dense packed cellulose has a heluva lot more thermal mass than rock wool at any given R-value, delaying the temperature peaks/troughs transferring through the assembly. This isn't particularly relevant from a fire performance point of view, but it can be from an actual performance vs. R-value perspective.

    Cellulose is substantially greener than rock wool- it is ~35% (sequestered) carbon by weight, with a very low processing-energy use greenhouse gas footprint.

    Cellulose protects structural wood from moisture far better than mineral wool or fiberglass ever can.

    On the down side, after a serious house fire is extinguished any cellulose that was fire-roasted long enough to actually ignite during the fire can smolder, and will require a bit more watching in the hours (or even a day or two) post fire than a mineral wool insulated house. If the initial fire didn't spread to adjoining houses it's unlikely any smoldering re-ignition would, but it's important to deal with it in the rare instances that it happens.

    I personally don't believe the (low) flammability of cellulose under extreme prolonged intense heat matters AT ALL in a "Dense urban environment". If it was a building with lots of high heat sources (planning on operating an iron foundry in the kitchen? :-) ) it would be relevant, but in those use-cases wood framing would be out as well.

    1. woobagoobaa | | #10

      Thank you Dana. We have control over the siding and interior insulation. Exterior insulation is not in the budget. Existing board sheathing stays in place.

      I will do a modified wicking test for the three mineral wool samples. I'll also include cellulose for comparison. Until them, I remain baffled re: why the batted mineral wool floats and the dense pack mineral wool sinks.

      1. maine_tyler | | #11

        If the stone wool material itself is denser than water, and the air spaces between are allowed to be filled with water, it stands to reason it would sink once enough air space has been displaced by water to overcome the buoyant effect. This still doesn't mean the material itself is hydrophilic or hydroscopic.

        But I understand how sinking insulation would give it the appearance of being not very moisture repellant. How long did it take to sink? I imagine the 'loose fill' nature of it makes the air spaces more readily accessible to water molecules.
        Any results from the wicking test suggested by Bill?

        1. Expert Member
          MALCOLM TAYLOR | | #12

          I'm trying to think of any other material that goes into a house you would pick based on its buoyancy.

          1. Expert Member
            BILL WICHERS | | #13

            Perhaps if you're building a house boat? hehe :-)

            Regarding the mineral wool, it's really about how much water it will hold after the source of moisture has been removed. Fiberglass is like a sponge, and stays soggy. Mineral wool can get saturated in a flood, but will rapidly shed that water and dry out after the flood waters recede. Mineral wool won't wick water, either, the water wont "rise" up a piece of mineral wool that is submerged only on the bottom. Fiberglass will 'suck' water up, similar to what happens if you hold a towel with the bottom dangling in water.

            If you have no control over the exterior sheathing, and you want to up your fire resistance, use all 5/8" type X drywall. That will do a lot more for fire resistance than the type of batt you stuff between the studs. Fire stop all penetrations between spaces too, basically build things commercial-style. "Fire resistance" isn't just about picking materials that are less flammable, it's also about limiting the ability of a fire to spready through the structure. A lot of the fire codes in the commerical world are about limiting the spread of fire. Even fire sprinklers act to keep the fire from spreading -- their purposes is not to put out a fire the way most people think, fire sprinklers in commercial buildings are intended to keep a fire contained until the fire department can get there to actually put the fire out.


      2. Expert Member
        Dana Dorsett | | #15

        >"why the batted mineral wool floats and the dense pack mineral wool sinks."

        Best guess: Binders alter and enhance the hygrophobic characteristics of the product. Batts have binders, blown mineral wool doesn't.

  7. maine_tyler | | #9

    I think generally a hydrophobic material won't 'seek equilibrium' with relative humidity conditions. Whether or not the air spaces collect water when submerged or via capillary is another story.

  8. woobagoobaa | | #16

    Revised test. Insulation samples dense packed into modified plastic bottles with holes in the bottom. All samples placed in a tray of water. Objective is to test relative wicking characteristics. Is this a reasonable test setup for testing hydrophic traits of the materials?

    Left to right: ThermaFibre InSUL FIL, ThermaFibre SAFB, Safe'n'Sound, cellulose. Will let them sit overnight.

    1. Expert Member
      BILL WICHERS | | #17

      Probably not, since you'll have some surface tension interaction on the inside surface of those plastic bottles. You'd be better off using loose samples, or using something that water won't cling to like polyethylene -- but that won't be a clear material. You may want to consider adding some fiberglass to your test too, since mineral wool is most commonly compared to fiberglass for this sort of thing.

      BTW, why are you so interested in testing this?


      1. woobagoobaa | | #18

        I don't want a sponge in my walls. And so far I am seeing substantial difference in the wetting behavior of batt versus dense pack mineral wool.

        1. Expert Member
          NICK KEENAN | | #19

          If you have liquid water in your walls you have other problems that you have to solve.

          1. woobagoobaa | | #20

            Agreed. I still don't want a sponge in my walls. Long term durability. Failure tolerant walls. This resonates ...

          2. Expert Member
            MALCOLM TAYLOR | | #21

            I agree with DC. I don't worry about bulk water intrusion into my exterior walls because I a) detail my walls properly to keep it out and b) don't run plumbing in them.

            I'm not arguing against mineral wool. If you follow Jon's frequent advice that we should design walls so that the exterior is much more permeable than the inside, I don't think it matters much what type of insulation you use.

  9. woobagoobaa | | #22

    My reno satisfies all the points you mentioned. I believe I have a perm to the exterior of at least 30 (see #3). We differ in that I am assuming I will have a WRB or flashing failure at some point, for some reason. Thus my strong preference that my insulation not act like a sponge.

    1. Expert Member
      MALCOLM TAYLOR | | #23

      Sounds like you have a good plan. I think most of the arguments have been about the usefulness of testing the mineral wool, not whether it's a good idea to use it.

      I can't argue against assuming some failure in the future, however unlikely it may be, as I am a pretty risk adverse builder too. Making decisions the help you sleep soundly seem like good ideas to me.

    2. Expert Member
      BILL WICHERS | | #24

      You should also test drying ability then. Take the mineral wool out of the water, see what happens. It should dry quickly compared to the others.

      I don't think you'll find any non-rigid insulation that is completely water proof. Something that dries quickly would be good if you're worried about water. I agree with Malcolm and DC -- detail things well to minimize the chances of water getting in to begin with. If your insulation can dry quick, then you'll have no issues after fixing whatever leak you have down the road. While you have a leak, any fibrous insulation is a problem since bulk water can still flow through it.


  10. woobagoobaa | | #25

    Here are my 24 hour wicking test results (thanks for all your counsel).

    From least to most wicking ...

    1. Fiberglass batt (did not absorb beyond the water level of the tray)
    2. ThermaFiber SAFB mineral wool batt. Minimal wicking.
    3. Safe'n'Sound mineral wool batt. Minimal wicking.
    4. Cellulose. Wicked right to the top of the sample.
    5. ThermaFiber INSUL FIL mineral wool. Wicked right to the top of the sample.

    Surprises? Fiberglass batt did the best. The loose fill mineral wool tied for worse with cellulose.

    I remain mystified re: any claim that the INSUL FIL is hydrophobic.

    I'll now do a drying test.

    1. Expert Member
      BILL WICHERS | | #26

      You might try calling the manufacturer. If something is messed up with their manufacturing process and you got a bad batch somehow, they'll probably want to know about it.


    2. maine_tyler | | #27

      If you take some of the loose fill mineral wool and pour water over a packed little clump of it, does the water bead up at all, or spread out/immediately absorb? Or if you push a clump briefly underwater, does there appear to be any air film surrounding the clump, even for a moment?

      I would second that drying capacity is likely the more relevant characteristic since this is cavity insulation without a drainage plane, (even of it sheds water, where's it going?). Whether it's a sponge or not, the water will be in your wall. It might depend a bit on the characteristic of the leak, but if its truly leaking INTO the wall, I don't really see how hydrophobia benefits. Redistribution could actually benefit (perhaps protect the bottom of the wall where it would collect if allowed to run straight through?).

      1. woobagoobaa | | #28

        No beading on the loose fill rock wool, batts bead as expected. Drying test of the samples continues.

        BTW I've been educated that I should be using the term rock or slag wool when referring to the Thermafiber or Rockwool products. Apologies for the confusion.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |