Image device and methods of forming the same

A method of forming an image sensor device, the method comprising:

• etching a plurality of first trenches in a periphery region of a substrate, the substrate having a front surface and a back surface, each of the first trenches having a depth D1 into the substrate;
• forming a mask layer over the front surface of the substrate, the mask layer having a plurality of openings in a pixel region of the substrate;
• narrowing each opening in the mask layer by forming a spacer on an interior surface of each opening in the mask layer;
• etching a plurality of second trenches through each narrowed opening in the pixel region, each of the second trenches having a depth D2 into the substrate, wherein the depth D1 is larger than the depth D2;
• forming at least one photodetector in the substrate of the pixel region, wherein the at least one photodetector is surrounded by the plurality of second trenches;
• and
• forming a color filter and a lens over the back surface of the substrate, wherein the color filter and the lens are aligned with the at least one photodetector.

An image sensor device, comprising:

• a substrate having a pixel region and a periphery region;
• a first mask layer over the substrate, the first mask layer having a plurality of first openings in the pixel region, wherein each first opening has a first width;
• a second mask layer over the first mask layer, the second mask layer having a plurality of second openings, wherein each second opening has a second width, and the second width is greater than the first width;
• a spacer layer over the first mask layer, the spacer layer having a first portion on a first interior sidewall surface of each second opening and a second portion on a second interior sidewall surface of each second opening, wherein a space between the first portion of the spacer layer and the second portion of the spacer layer is about equal to the first width;
• a plurality of isolation features in the periphery region, wherein each isolation feature has a depth D1;
• a plurality of trenches in the pixel region, wherein each trench has a depth D2, and the depth D1 is larger than the depth D2;
• and
• at least one light-sensing region in the pixel region, the light-sensing region being surrounded by the trenches.