Advanced styling effects

box-shadow allows you to apply one or more drop shadows to an element's box. Like text shadows, box shadows are supported pretty well across browsers, including IE9+ and Edge. Users of older IE versions might just have to cope with no shadows, so just test your designs to make sure your content is legible without them.

You can find the examples in this section at box-shadow.html (see the source code too).

Let's look at a simple example to get things started. First, some HTML:

<article class="simple">
  <p>
    <strong>Warning</strong>: The thermostat on the cosmic transcender has
    reached a critical level.
  </p>
</article>

Now the CSS:

p {
  margin: 0;
}

article {
  max-width: 500px;
  padding: 10px;
  background-color: red;
  background-image: linear-gradient(
    to bottom,
    rgb(0 0 0 / 0%),
    rgb(0 0 0 / 25%)
  );
}

.simple {
  box-shadow: 5px 5px 5px rgb(0 0 0 / 70%);
}

This gives us the following result:

You'll see that we've got four items in the box-shadow property value:

1. The first length value is the horizontal offset — the distance to the right the shadow is offset from the original box (or left, if the value is negative).
2. The second length value is the vertical offset — the distance downwards that the shadow is offset from the original box (or upwards, if the value is negative).
3. The third length value is the blur radius — the amount of blurring applied to the shadow.
4. The color value is the base color of the shadow.

You can use any length and color units that would make sense to do so to define these values.

You can also specify multiple box shadows in a single box-shadow declaration, by separating them with commas:

<article class="multiple">
  <p>
    <strong>Warning</strong>: The thermostat on the cosmic transcender has
    reached a critical level.
  </p>
</article>

p {
  margin: 0;
}

article {
  max-width: 500px;
  padding: 10px;
  background-color: red;
  background-image: linear-gradient(
    to bottom,
    rgb(0 0 0 / 0%),
    rgb(0 0 0 / 25%)
  );
}

.multiple {
  box-shadow: 1px 1px 1px black,
              2px 2px 1px black,
              3px 3px 1px red,
              4px 4px 1px red,
              5px 5px 1px black,
              6px 6px 1px black;
}

Now we get this result:

We've done something fun here by creating a raised box with multiple colored layers, but you could use it in any way you want, for example to create a more realistic look with shadows based on multiple light sources.

Unlike text-shadow, box-shadow has an inset keyword available — putting this at the start of a shadow declaration causes it to become an inner shadow, rather than an outer shadow. Let's have a look and see what we mean.

First, we'll go with some different HTML for this example:

<button>Press me!</button>

button {
  width: 150px;
  font-size: 1.1rem;
  line-height: 2;
  border-radius: 10px;
  border: none;
  background-image: linear-gradient(to bottom right, #777, #ddd);
  box-shadow:
    1px 1px 1px black,
    inset 2px 3px 5px rgb(0 0 0 / 30%),
    inset -2px -3px 5px rgb(255 255 255 / 50%);
}

button:focus,
button:hover {
  background-image: linear-gradient(to bottom right, #888, #eee);
}

button:active {
  box-shadow:
    inset 2px 2px 1px black,
    inset 2px 3px 5px rgb(0 0 0 / 30%),
    inset -2px -3px 5px rgb(255 255 255 / 50%);
}

This gives us the following result:

Here we've set up some button styling along with focus/hover/active states. The button has a simple black box shadow set on it by default, plus a couple of inset shadows, one light and one dark, placed on opposite corners of the button to give it a nice shading effect.

When the button is pressed in, the active state causes the first box shadow to be swapped for a very dark inset shadow, giving the appearance of the button being pressed in.

While you can't change the composure of an image using CSS, there are some creative things you can do. One very nice property, which can help you bring interest to your designs, is the filter property. This property enables Photoshop-like filters right from CSS.

In the example below we have used two different values for filter. The first is blur() — this function can be passed a value to indicate how much the image should be blurred.

The second is grayscale(); by using a percentage we are setting how much color we want to be removed.

Play with the percentage and pixel parameters in the live example to see how the images change. You could also swap the values for some others. Try contrast(200%), invert(100%) or hue-rotate(20deg) on the live example above. Take a look at the MDN page for filter for many other options you could try.

You can apply filters to any element and not just images. Some of the filter options available do very similar things to other CSS features, for example drop-shadow() works in a very similar way and gives a similar effect to box-shadow or text-shadow. The really nice thing about filters however, is that they work on the exact shapes of the content inside the box, not just the box itself as one big chunk, so it is worth knowing the difference.

In this next example we are applying our filter to a box, and comparing it to a box shadow. As you can see, the drop-shadow filter follows the exact shape of the text and border dashes. The box shadow just follows the square of the box.

CSS blend modes allow us to add blend modes to elements that specify a blending effect when two elements overlap — the final color shown for each pixel will be the result of a combination of the original pixel color, and that of the pixel in the layer underneath it. Blend modes are again very familiar to users of graphics applications like Photoshop.

There are two properties that use blend modes in CSS:

• background-blend-mode, which blends together multiple background images and colors set on a single element.
• mix-blend-mode, which blends together the element it is set on with elements it is overlapping — both background and content.

You can find a lot more examples than are available here in our blend-modes.html example page (see source code), and on the <blend-mode> reference page.

Again, let's look at some examples so we can understand this better. First, background-blend-mode — here we'll show a couple of simple <div>s, so you can compare the original with the blended version:

<div></div>
<div class="multiply"></div>

Now some CSS — we are adding to the <div> one background image and a green background color:

div {
  width: 250px;
  height: 130px;
  padding: 10px;
  margin: 10px;
  display: inline-block;
  background: url(colorful-heart.png) no-repeat center 20px;
  background-color: green;
}

.multiply {
  background-blend-mode: multiply;
}

The result we get is this — you can see the original on the left, and the multiply blend mode on the right:

Now let's look at mix-blend-mode. Here we'll present the same two <div>s, but each one is now sat on top of a simple <div> with a purple background, to show how the elements will blend together:

<article>
  No mix blend mode
  <div></div>
  <div></div>
</article>

<article>
  Multiply mix
  <div class="multiply-mix"></div>
  <div></div>
</article>

Here's the CSS we'll style this with:

article {
  width: 280px;
  height: 180px;
  margin: 10px;
  position: relative;
  display: inline-block;
}

div {
  width: 250px;
  height: 130px;
  padding: 10px;
  margin: 10px;
}

article div:first-child {
  position: absolute;
  top: 10px;
  left: 0;
  background: url(colorful-heart.png) no-repeat center 20px;
  background-color: green;
}

article div:last-child {
  background-color: purple;
  position: absolute;
  bottom: -10px;
  right: 0;
  z-index: -1;
}

.multiply-mix {
  mix-blend-mode: multiply;
}

This gives us the following results:

You can see here that the multiply mix blend has blended together not only the two background images, but also the color from the <div> below it too.

While it is true that everything in CSS is a rectangular box, and images are a physical rectangular box, we can make it look as if our content flows around non-rectangular things by using CSS Shapes.

The CSS Shapes specification enables the wrapping of text around a non-rectangular shape. It's especially useful when working with an image which has some white-space you might want to float text around.

In the image below we have a pleasingly round balloon. The actual file is rectangular, but by floating the image (shapes only apply to floated elements) and using the shape-outside property with a value of circle(50%), we can give the effect of the text following the line of the balloon.

The shape in this example is not reacting to the content of the image file. Instead, the circle function is taking its center point from the center of the image file, as if we had put a compass in the middle of the file and drawn a circle that fits inside the file. It is that circle that the text flows around.

The circle() function is just one of a few basic shapes that are defined, however there are a number of different ways to create shapes. For more information and example code for CSS Shapes see the Guides to CSS Shapes on MDN.

Another feature we thought we'd mention briefly is the text value for background-clip. When used along with the proprietary -webkit-text-fill-color: transparent; feature, this allows you to clip background images to the shape of the element's text, making for some nice effects. This is not an official standard, but has been implemented across multiple browsers, as it is popular, and used fairly widely by developers. When used in this context, both of the properties would require a -webkit- vendor prefix, even for Non-Webkit/Chrome-based browsers:

.text-clip {
  -webkit-background-clip: text;
  -webkit-text-fill-color: transparent;
}

So why have other browsers implemented a -webkit- prefix? Mainly for browser compatibility — so many web developers have started implementing websites with -webkit- prefixes that it started to look like the other browsers were broken, whereas in actual fact they were following the standards. So they were forced to implement a few such features. This highlights the danger of using non-standard and/or prefixed CSS features in your work — not only do they cause browser compatibility issues, but they are also subject to change, so your code could break at any time. It is much better to stick to the standards.

If you do want to use such features in your production work, make sure to test across browsers thoroughly and check that, where these features don't work, the site is still usable.

We hope this article was fun — playing with shiny toys generally is, and it is always interesting to see what kinds of advanced styling tools are becoming available in modern browsers.

• Previous
• Overview: CSS building blocks
• Next